From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 4 Nov 1997 15:12:03 -0800
From: "Luiz Otavio L. Zorzella" <zorzella@laser.cps.softex.br>
To: flexfax@sgi.com
Subject: flexfax: sharing a modem with pppd
Sender: owner-flexfax@celestial.com


Hi,

I'd like to know if it's possible in hylafax to send a fax through the
modem being used by pppd.

I mean: if I can turn of pppd, send a fax, and turn pppd on again.

I tryied doing this, but seems that the hylafax server insists on
resolving my machine's name when it gets a requisition, so that, when
I try to use "sendfax" when pppd is off, I'll get a message:


Service not available, remote server closed connection


And in the syslog, I have the message:

Nov  4 15:09:30 KH HylaFAX[1041]: CAUGHT SIGNAL 11

Any clues?

Thanks.

Z.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
X-Authentication-Warning: tui.pinnacle.co.nz: jonc owned process doing -bs
Date: Thu, 6 Nov 1997 09:05:13 +1300 (NZDT)
From: Jonathan Chen <jonc@pinnacle.co.nz>
To: "Luiz Otavio L. Zorzella" <zorzella@laser.cps.softex.br>
cc: flexfax@sgi.com
Subject: Re: flexfax: sharing a modem with pppd
Sender: owner-flexfax@celestial.com

On Tue, 4 Nov 1997, Luiz Otavio L. Zorzella wrote:

> 
> Hi,
> 
> I'd like to know if it's possible in hylafax to send a fax through the
> modem being used by pppd.
> 
> I mean: if I can turn of pppd, send a fax, and turn pppd on again.

It should be possible to keep pppd running, as well as having HylaFAX
staying around.

> I tryied doing this, but seems that the hylafax server insists on
> resolving my machine's name when it gets a requisition,

Looks like you may have to set up a DNS or check your current DNS
queries when pppd is running. Something pppd is doing is causing
HylaFAX's DNS queries to fail - significantly.

Cheers.
--
Jonathan Chen <jonc@pinnacle.co.nz>                 Once is dumb luck.
                                                 Twice is coincidence.
             Three times and Somebody Is Trying To Tell You Something.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 5 Nov 1997 11:24:24 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: "Luiz Otavio L. Zorzella" <zorzella@laser.cps.softex.br>
Cc: flexfax@sgi.com
Subject: Re: flexfax: sharing a modem with pppd
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Tue, 4 Nov 1997, Luiz Otavio L. Zorzella wrote:

> Hi,
> 
> I'd like to know if it's possible in hylafax to send a fax through the
> modem being used by pppd.
> 
> I mean: if I can turn of pppd, send a fax, and turn pppd on again.
> 
> I tryied doing this, but seems that the hylafax server insists on
> resolving my machine's name when it gets a requisition, so that, when
> I try to use "sendfax" when pppd is off, I'll get a message:
> 
> 
> Service not available, remote server closed connection

Hmm. Sounds like PPP and HylaFAX are not (directly) the problem, but
rather your domain name resolution when you don't have PPP running. Do
you have a default fax server set up with FAXSERVER somewhere? Try
killing the PPP, and seeing if you can run "faxstat -v".

If that doesn't work, try setting your FAXSERVER to "localhost" and see
what you get....

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNGCdsT/+ItycgIJRAQFdLAP/Vrqxm4QSjcYCw1NVAJaJXEyoO+tf6Je1
ojzahKo/YwSRdYgLHAg/zojcAh7bVo4lUvVGWUjQcpo0AzypJil6fadSaNBJFLQ/
eOTr9A+zeglATOUa/VY3m9NyA3P9LtKFGWnIi5Zn4AJxB2s0F0uKUy3FvD8icl5R
U830WebxCt8=
=OtaF
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: sharing a modem with pppd
To: raoul@cirl.meei.harvard.edu (Nico Garcia)
Date: Fri, 7 Nov 1997 08:07:09 +0100 (MEZ)
Cc: zorzella@laser.cps.softex.br, flexfax@sgi.com, guru@thias.muc.de
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Nico Garcia wrote:
    
    On Tue, 4 Nov 1997, Luiz Otavio L. Zorzella wrote:
    
    > Hi,
    > 
    > I'd like to know if it's possible in hylafax to send a fax through the
    > modem being used by pppd.
    > 
    > I mean: if I can turn of pppd, send a fax, and turn pppd on again.
    > 
    > I tryied doing this, but seems that the hylafax server insists on
    > resolving my machine's name when it gets a requisition, so that, when
    > I try to use "sendfax" when pppd is off, I'll get a message:
    > 
    > 
    > Service not available, remote server closed connection
    
    Hmm. Sounds like PPP and HylaFAX are not (directly) the problem, but
    rather your domain name resolution when you don't have PPP running. Do
    you have a default fax server set up with FAXSERVER somewhere? Try
    killing the PPP, and seeing if you can run "faxstat -v".
    
    If that doesn't work, try setting your FAXSERVER to "localhost" and see
    what you get....
    
    			Nico Garcia
    			Engineer, CIRL 
    			Mass. Eye and Ear Infirmary
    			raoul@cirl.meei.harvard.edu

The FAXSERVER environment variable controls which host and port
the clients (e.g. faxstat(1) or sendfax(1)) should contact
for the "hylafax" service of hfaxd(1M). This variable has
nothing to do with the running hfaxd(1M) and if I understand
the problem report of Luiz Otavio right the problem is that
a client *can* reach the service at the host, the hfaxd(1M)
gets the connection, forks and aborts with SIG 11 in
case the PPP-link isn't up. This sounds like related to a name
resolving problem but is unrelated to FAXSERVER setting.

At the moment I don't know which OS type Luiz Otavio is
running.... ah, it's Linux (just checked my HylaFAX mail
folders). Luiz Otavio, please run a strace(1) command
with the -f flag (to monitor also forked childs) against
the running hfaxd(1M), start this before you fire up the
client (or just use "telnet localhost hylafax") and
send me the output of the strace(1). I'll also send
you a patch (directly) which deals with this problem in pl2
(but without the strace I'm not sure if you hit the same
situation).

I also managed to compile HylaFAX at home on the S.u.S.E. Linux
box of my son (normaly I run SVR4.2) and can now answer
Linux related questions better than before.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 7 Nov 1997 10:54:00 -0500 (EST)
From: Helen Birkmann <hbirkmann@mkl.com>
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
cc: "Luiz Otavio L. Zorzella" <zorzella@laser.cps.softex.br>, flexfax@sgi.com
Subject: Re: flexfax: sharing a modem with pppd
Sender: owner-flexfax@celestial.com

On Wed, 5 Nov 1997, Nico Garcia wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> 
> On Tue, 4 Nov 1997, Luiz Otavio L. Zorzella wrote:
> 
> > Hi,
> > 
> > I'd like to know if it's possible in hylafax to send a fax through the
> > modem being used by pppd.
> > 
> > I mean: if I can turn of pppd, send a fax, and turn pppd on again.
> > 
> > I tryied doing this, but seems that the hylafax server insists on
> > resolving my machine's name when it gets a requisition, so that, when
> > I try to use "sendfax" when pppd is off, I'll get a message:
> > 
> > 
> > Service not available, remote server closed connection
> 
> Hmm. Sounds like PPP and HylaFAX are not (directly) the problem, but
> rather your domain name resolution when you don't have PPP running. Do
> you have a default fax server set up with FAXSERVER somewhere? Try
> killing the PPP, and seeing if you can run "faxstat -v".
> 
> If that doesn't work, try setting your FAXSERVER to "localhost" and see
> what you get....
> 
> 			Nico Garcia
> 			Engineer, CIRL 
> 			Mass. Eye and Ear Infirmary
> 			raoul@cirl.meei.harvard.edu
> 
> -----BEGIN PGP SIGNATURE-----
> Version: 2.6.2
> 
> iQCVAwUBNGCdsT/+ItycgIJRAQFdLAP/Vrqxm4QSjcYCw1NVAJaJXEyoO+tf6Je1
> ojzahKo/YwSRdYgLHAg/zojcAh7bVo4lUvVGWUjQcpo0AzypJil6fadSaNBJFLQ/
> eOTr9A+zeglATOUa/VY3m9NyA3P9LtKFGWnIi5Zn4AJxB2s0F0uKUy3FvD8icl5R
> U830WebxCt8=
> =OtaF
> -----END PGP SIGNATURE-----
> 
> 

We are using BSDI 3.0 and the instructions read specifically that the
getty daemon should NOT be used on the same modem that you are setting up
for hylafax.

Regards
Helen
************************************************************************
*	MKL.Net		        	Internet Service Provider      *
*	A service of CPU Industries                                    *
*	Bainbridge, New York, USA      	PPP, dial-up, telnet, uucp,    *
*	http://www.mkl.com		virtual domains, web design    *
*	https://mall.mkl.com	Secured server - Merchant services     *
************************************************************************

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Steve Williams <steve@genie96.com>
Subject: Re: flexfax: Okay, so it was a FAQ...
To: elliots@epower.net (Elliot Smorodinsky)
Date: Thu, 11 Sep 1997 16:39:21 -0600 (MDT)
Cc: flexfax@sgi.com
Read-Receipt-To: steve@genie96.com
Sender: owner-flexfax@celestial.com

> 
> 	In my earlier post, I asked about making HylaFax work with tip. 
> Afterwards, I consulted the FAQ and came away feeling much embarrassed, 
> but not at all enlightened.
> 
> 	Once again: server is OSF 4.0, Hylafax 4.0pl1.
> 
> Quoting from the FAQ:
> 
> >...First verify that your data communication software is configured to use 
> >the correct modem initialization strings
> 	They do. 
> 
> > and that both HylaFAX
> > and the program(s) in question are using the same device name.
> 	They are.
> 
> 	To clarify, tip works. Tip works very well. So does HylaFax. My 
> problem, specifically, is that getty recognizes the lock file and obeys 
> it... for about 30 seconds. Then, as the modem is in the midst of 
> dialing, getty calmly declares the UUCP lock file 'stale' and blows it away,
> grabbing control of the modem back from 'tip'. And I cannot for the life of
> me find the place in the config files where that 30-second default resides...
> 

Hi,

Just so you understand the UUCP locking mechanism a bit better...

The UUCP Lockfile contains the Process ID of the process that contains the
lockfile.

There are two formats for lockfiles.
	- Binary Lockfiles, where the PID is stored in a binary format
	- Ascii Lockfiles, where the PID is stored in an ascii format,
	  human readable form.

Before ANY program nukes the lockfile, the program checks to see if the
Process still exists.  If the process still exists, the Lockfile is considered
valid, and NOT removed.  The process is checked for existance by using the 
"kill 0" signal on the PID.  This even works from the shell.

It is critical that two criteria are met for the locking to work properly.
	- All programs on a system accessing the modem MUST use the same format
	  of UUCP Lockfile.  ( Binary OR Ascii )

	- All programs on a system accessing the modem MUST be looking for the
	  lockfile in the same location ( directory ).

To diagnose the problem, try..

1.  Send a ( longish ) fax.  Once it is transmitting, do a "uustat -p"
    This should show some "fax" process that has the port.  If not, then
    HylaFAX is not using the same directory as your system's UUCP is expecting.
    OR HylaFAX is writing the wrong format lock file.

2.  While the fax is being sent, change into the UUCP lock directory.  It 
    could be /etc/locks, or /usr/lib/uucp, ..., and look for a file with the
    name LCK..[tty_name].

3.  "cat" this file.  If you can read it, you should see the process id of
    the sending fax program.  If you cannot read it ( or get garbage ), then
    HylaFAX is set to use binary lock files.  This is OK if the uustat -p
    showed the sending fax process, because your system uses binary lockfiles.

4.  Wait for the fax to finish.

5.  "tip" to the port.

6.  From another terminal ( or shell out, or whatever, but do it within 30 
    seconds <g> ), do a uustat -p.  If that shows "tip" as having the port,
    then I am really confused.  If it doesn't, then "tip" is either putting
    the lockfile in a different directory than is expected, OR it is using
    the wrong format lockfile.

7.  Change directory into the lock directory, and "cat" the lockfile to 
    determine what format of lockfile "tip" is writing.

Figure it out from there....,

I would suspect that HylaFAX is compiled up to use the wrong format lockfile
for your system.

Good Luck,
-- 
	Steve Williams, Calgary, Alberta, Canada
	Genie Computer Systems Inc.
	steve@genie96.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
To: sam@engr.sgi.com
cc: flexfax@sgi.com, Phil Abercrombie <phil@Comp.VUW.AC.NZ>
Subject: bug in 4.0pl1 with Letter PS on an A4 machine
Date: Tue, 26 Nov 1996 12:54:51 +0000
From: "Mr. Arlington Hewes" <tpcadmin@info.tpc.int>
Sender: owner-flexfax@celestial.com


Dear Sam (and list),


Problem:

Letter-formatted postscript output from Apple LW8.2 PS print driver will not 
FAX on an A4-configured HylaFAX server.


Symptoms:

Initial faulure to FAX with "Client does not support document page 
width, max remote page width 1728 pixels, image width 2383 pixels".  Then the 
job file gets corrupted, and the job gets "lost".  There is no notification 
until faxq is restarted and it noticed the corrupt job file and mails the 
originator who may have sent the FAX days earlier.


Syslog output (session traces are uninteresting - the coverpage is sent and 
then it fails with the message "Client does not support document page 
width, max remote page width 1728 pixels, image width 2383 pixels"):

Nov 22 00:28:59 fax-gw FaxSend[24621]: MODEM MULTI-TECH SYSTEMS MT1932ZDXK/0113
Nov 22 00:29:00 fax-gw FaxSend[24621]: SEND FAX: JOB 23385 DEST +441908265xxx 
COMMID 00001189
Nov 22 00:30:07 fax-gw FaxSend[24621]: Client does not support document page 
width, max remote page width 1728 pixels, image width 2383 pixels
Nov 22 00:30:09 fax-gw FaxQueuer[19451]: JOB 23385 (active dest +441908265xxx 
pri 126 tts 0:00 killtime 23:56:27): CMD DONE: exit status 0x300
Nov 22 00:30:10 fax-gw FaxQueuer[19451]: JOB 23385 (ready dest +441908265xxx 
pri 125 tts 0:00 killtime 23:56:26): SEND INCOMPLETE: retry immediately; 
Client does not support document page width, max remote page width 1728 
pixels, image width 2383 pixels
Nov 22 00:30:10 fax-gw FaxQueuer[19451]: JOB 23385 (active dest +441908265xxx 
pri 125 tts 0:00 killtime 23:56:26): PREPARE START
Nov 22 00:30:10 fax-gw FaxQueuer[24632]: JOB 23385 (active dest +441908265xxx 
pri 125 tts 0:00 killtime 23:56:26): CONVERT DOCUMENT: bin/ps2fax -o 
docq/doc20129.ps;00 -r 98 -w 1728 -l 297 -m 20 -1 docq/doc20129.ps.23385
Nov 22 00:30:13 fax-gw FaxQueuer[19451]: JOB 23385 (active dest +441908265xxx 
pri 125 tts 0:00 killtime 23:56:23): PREPARE DONE
Nov 22 00:30:13 fax-gw FaxQueuer[19451]: JOB 23385 (active dest +441908265xxx 
pri 125 tts 0:00 killtime 23:56:23): CMD START /usr/local/sbin/faxsend -m 
ttyC00 sendq/q23385 (PID 24637)
Nov 22 00:30:21 fax-gw FaxSend[24637]: MODEM MULTI-TECH SYSTEMS MT1932ZDXK/0113
Nov 22 00:30:21 fax-gw FaxSend[24637]: SEND FAX: JOB 23385 DEST +441908265xxx 
COMMID 00001190
Nov 22 00:31:10 fax-gw FaxQueuer[19451]: Bogus document handed to unrefDoc: 00
Nov 22 00:31:10 fax-gw FaxQueuer[19451]: Bogus document handed to unrefDoc: 00
Nov 22 00:31:10 fax-gw FaxSend[24637]: Unable to rename transmitted document 
docq/doc20128: Invalid argument
Nov 22 00:31:10 fax-gw FaxSend[24637]: Unable to rename transmitted document 
docq/doc20128: Invalid argument
Nov 22 00:31:10 fax-gw FaxSend[24637]: Client does not support document page 
width, max remote page width 1728 pixels, image width 2383 pixels
Nov 22 00:31:12 fax-gw FaxQueuer[19451]: JOB 23385 (active dest +441908265xxx 
pri 125 tts 0:00 killtime 23:55:24): CMD DONE: exit status 0x300
Nov 22 00:31:12 fax-gw FaxQueuer[19451]: JOB 23385 (ready dest +441908265xxx 
pri 124 tts 0:00 killtime 23:55:24): SEND INCOMPLETE: retry immediately; 
Client does not support document page width, max remote page width 1728 
pixels, image width 2383 pixels
Nov 22 00:31:12 fax-gw FaxQueuer[19451]: sendq/q23385: line 52: Can not access 
document file "docq/doc20128": No such file or directory
Nov 22 00:31:12 fax-gw FaxQueuer[19451]: sendq/q23385: line 52: Can not access 
document file "docq/doc20128": No such file or directory
Nov 22 00:31:12 fax-gw FaxQueuer[19451]: JOB 23385: Could not read job file
Nov 22 00:31:12 fax-gw FaxQueuer[19451]: JOB 23385: Could not read job file


I can reproduce the problem, as can Jonathan Chen in NZ.  Nico Garcia FAXes the postscript quite happily, but his server is set up for Letter.  I can supply the Postscript file to anyone who would like to test their A4 setups.  I can test patches . . . about all I cannot do is hand you all a patch.  But I think someone needs to look at this. . . the Apple Laserwriter is a pretty generic driver, and a lot of people use it to generate their PS.

=====
I have an even more detailed post, with detailed session logs and such, 
should anyone be intrigued enough to chase this one.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: What's "REMOTE HANGUP: DTE to DCE data underflow (code 43)"?
To: oyang@phoebe.fcit.monash.edu.au (Kai O'Yang)
Date: Tue, 23 Sep 1997 08:33:28 +0100 (BST)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> Sep 23 13:16:18.62: [15317]: REMOTE HANGUP: DTE to DCE data underflow (code 43)

Without the document specifying the codes, I can only go on the wording,
which appears to say that the remote fax software failed to respond to the
remote faxmodem in time.  So unless the wording is totally misleading,
the problem is not yours.

It is possible that the REMOTE is incorrect in the above, in which case the
problem would be that your machine is too slow for the modem speed negotiated,
or your computer to modem interface speed is too slow.  You cut out the part
of the trace that might give clues about this.

DCE = modem
DTE ~= computer
underflow = underrun = data didn't arrive in time to meet a critical
timing requirement.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 24 Sep 97 14:04:43 EDT
From: Glenn Burkhardt <glenn@aoi.ultranet.com>
To: andy@wx.gtegsc.com
Cc: flexfax@fddi-sgi.engr.sgi.com
Subject: flexfax: suggestions for changes to tkhylafax-3.0
Sender: owner-flexfax@celestial.com

I'm attaching a patch file for some improvements to tkhylafax.

1.  The fields that pick up the comments, etc., for the cover sheet
wouldn't correctly pass the characters ['"$\] to the cover sheet
program, so, for example, if you wanted to include a price with a $
sign, you couldn't.

2.  I've added a system-wide default resource file, to be put in the
directory '/usr/local/lib/tkhylafax-3.0, or wherever the TCL source
lives.  It's name is 'tkhylafax-defaults'.

3.  I've added a hook to pick up the invocation of 'wish', and used the
value in the Makefile to change it.  The change to the Makefile also
cleans up the patch of the 'tkhylafax' main source file, but REQUIRES
that the name of the original file be changed to 'tkhylafax.unpatched',
i.e, after applying the patch file, you need to 

	mv tkhylafax tkhylafax.unpatched

Thanks for putting this program together, and I hope you find these
changes useful.



*** orig/Makefile	Wed Sep 24 13:32:12 1997
--- Makefile	Wed Sep 24 13:31:03 1997
***************
*** 60,75 ****
  	if [ ! -d $(TCLPRG_DEST) ]; then mkdir -p $(TCLPRG_DEST); fi
  	chmod 775 $(TCLPRG_DEST)
  
! tkhylafax: $(TCLPRG_DEST)
  	@echo "Patching tkhylafax..." 
- 	mv tkhylafax tkhylafax.unpatched
  	sed -e 's|XXX_TKHYLAFAX_XXX|$(TCLLIB_DEST)|g' \
! 	    tkhylafax.unpatched > tkhylafax
  	@echo "Done" 
  	@echo "Installing tkhylafax command..." 
  	cp $(TCLPRG_FILE) $(TCLPRG_DEST)/$@
  	chmod 775 $(TCLPRG_DEST)/$@
- 	mv tkhylafax.unpatched tkhylafax
  	@echo "Done" 
  
  install.man:
--- 60,74 ----
  	if [ ! -d $(TCLPRG_DEST) ]; then mkdir -p $(TCLPRG_DEST); fi
  	chmod 775 $(TCLPRG_DEST)
  
! tkhylafax: $(TCLPRG_DEST) Makefile
  	@echo "Patching tkhylafax..." 
  	sed -e 's|XXX_TKHYLAFAX_XXX|$(TCLLIB_DEST)|g' \
! 	    -e 's|XXX_WISH_XXX|$(WISH)|g' \
! 	    tkhylafax.unpatched > $(TCLPRG_FILE)
  	@echo "Done" 
  	@echo "Installing tkhylafax command..." 
  	cp $(TCLPRG_FILE) $(TCLPRG_DEST)/$@
  	chmod 775 $(TCLPRG_DEST)/$@
  	@echo "Done" 
  
  install.man:
*** orig/preferences.t	Wed Sep 24 13:38:20 1997
--- preferences.t	Wed Sep 24 13:38:57 1997
***************
*** 36,46 ****
  # Set user preferences
  #
  
! proc initPreference { {filename {}} } {
  
      global env
  
!     if {$filename != ""} {
          set database $filename
      } elseif {[info exists env(TKHYLAFAX_DEFAULTS)]} {
          set database $env(TKHYLAFAX_DEFAULTS)
--- 36,46 ----
  # Set user preferences
  #
  
! proc initPreference { filename } {
  
      global env
  
!     if {$filename != "" && [file exists $filename]} {
          set database $filename
      } elseif {[info exists env(TKHYLAFAX_DEFAULTS)]} {
          set database $env(TKHYLAFAX_DEFAULTS)
*** orig/faxIt.t	Wed Sep 24 13:40:31 1997
--- faxIt.t	Wed Sep 24 13:42:38 1997
***************
*** 67,76 ****
  
      if {[string length $Th_Info(name)]} {
          # This MUST appear.
!         append options " -d \"$Th_Info(name)@$Th_Info(phone)\""
  
          # Make an ID string for later usage
!         append options " -i \"$Th_Info(name)@$Th_Info(company)\"" 
      } else {
          # This MUST appear.
          append options " -d $Th_Info(phone)"
--- 67,78 ----
  
      if {[string length $Th_Info(name)]} {
          # This MUST appear.
! 	regsub -all {["'$\\]} $Th_Info(name) {\\&} name
!         append options " -d \"$name@$Th_Info(phone)\""
  
          # Make an ID string for later usage
! 	regsub -all {["'$\\]} $Th_Info(company) {\\&} company
!         append options " -i \"$name@$company\"" 
      } else {
          # This MUST appear.
          append options " -d $Th_Info(phone)"
***************
*** 91,115 ****
  	# Make command. Ignore the following options: -k -s -h -v 
  	append faxcmd "sendfax "
  
!          foreach ln [lsort [array name Comments]] { 
  	    append comments "$Comments($ln) "
!          }    
  
  	set comments [string trim $comments]
  	if {[string length $comments]} {
  	    append faxcmd " -c \"$comments\""
  	}
  
  	if {[string length $Th_Info(regard)]} {
! 	    append faxcmd " -r \"$Th_Info(regard)\""
  	}
  
  	if {[string length $Th_Info(company)]} {
! 	    append faxcmd " -x \"$Th_Info(company)\""
  	}
  
  	if {[string length $Th_Info(location)]} {
! 	    append faxcmd " -y \"$Th_Info(location)\""
  	}
  
  	append faxcmd $options
--- 93,121 ----
  	# Make command. Ignore the following options: -k -s -h -v 
  	append faxcmd "sendfax "
  
!         foreach ln [lsort [array name Comments]] { 
  	    append comments "$Comments($ln) "
!         }    
  
  	set comments [string trim $comments]
  	if {[string length $comments]} {
+ 	    regsub -all {["'$\\]} $comments {\\&} comments
  	    append faxcmd " -c \"$comments\""
  	}
  
  	if {[string length $Th_Info(regard)]} {
! 	    regsub -all {["'$\\]} $Th_Info(regard) {\\&} regarding
! 	    append faxcmd " -r \"$regarding\""
  	}
  
  	if {[string length $Th_Info(company)]} {
! 	    regsub -all {["'$\\]} $Th_Info(company) {\\&} company
! 	    append faxcmd " -x \"$company\""
  	}
  
  	if {[string length $Th_Info(location)]} {
! 	    regsub -all {["'$\\]} $Th_Info(location) {\\&} location
! 	    append faxcmd " -y \"$location\""
  	}
  
  	append faxcmd $options
*** orig/mkCover.t	Wed Sep 24 13:40:31 1997
--- mkCover.t	Wed Sep 24 13:42:38 1997
***************
*** 52,74 ****
  
      set comments [string trim $comments]
      if {[string length $comments]} {
  	append faxcmd " -c \"$comments\""
      }
  
      if {[string length $Th_Info(regard)]} {
! 	append faxcmd " -r \"$Th_Info(regard)\""
      }
  
      if {[string length $Th_Info(company)]} {
! 	append faxcmd " -x \"$Th_Info(company)\""
      }
  
      if {[string length $Th_Info(location)]} {
! 	append faxcmd " -l \"$Th_Info(location)\""
      }
  
      if {[string length $Th_Info(name)]} {
! 	append faxcmd " -t \"$Th_Info(name)\""
      } else {
  	append faxcmd " -t \"Unknown\""  
      }
--- 52,79 ----
  
      set comments [string trim $comments]
      if {[string length $comments]} {
+ 	regsub -all {["'$\\]} $comments {\\&} comments
  	append faxcmd " -c \"$comments\""
      }
  
      if {[string length $Th_Info(regard)]} {
! 	regsub -all {["'$\\]} $Th_Info(regard) {\\&} regarding
! 	append faxcmd " -r \"$regarding\""
      }
  
      if {[string length $Th_Info(company)]} {
! 	regsub -all {["'$\\]} $Th_Info(company) {\\&} company
! 	append faxcmd " -x \"$company\""
      }
  
      if {[string length $Th_Info(location)]} {
! 	regsub -all {["'$\\]} $Th_Info(location) {\\&} location
! 	append faxcmd " -l \"$location\""
      }
  
      if {[string length $Th_Info(name)]} {
! 	regsub -all {["'$\\]} $Th_Info(name) {\\&} name
! 	append faxcmd " -t \"$name\""
      } else {
  	append faxcmd " -t \"Unknown\""  
      }
***************
*** 122,128 ****
  	}
      } else {
  	if {[string length $Th_Info(from)]} {
! 	    append faxcmd " -f \"$Th_Info(from)\""
  	} else {
  	    append faxcmd " -f \"Unknown\""
  	}
--- 127,134 ----
  	}
      } else {
  	if {[string length $Th_Info(from)]} {
! 	    regsub -all {["'$\\]} $Th_Info(from) {\\&} from
! 	    append faxcmd " -f \"$from\""
  	} else {
  	    append faxcmd " -f \"Unknown\""
  	}
*** orig/tkhylafax	Wed Sep 24 13:34:59 1997
--- tkhylafax	Wed Sep 24 13:46:00 1997
***************
*** 1,6 ****
  #!/bin/sh
  #\
! exec wish "$0" ${1+"$@"}
  
  #
  # 
--- 1,6 ----
  #!/bin/sh
  #\
! exec XXX_WISH_XXX "$0" ${1+"$@"}
  
  #
  # 
***************
*** 349,355 ****
  }
  
  lappend auto_path $Th_Info(libraryPath)
! initPreference
  setOptions
  initRolodex
  initBatch
--- 349,355 ----
  }
  
  lappend auto_path $Th_Info(libraryPath)
! initPreference $Th_Info(libraryPath)/tkhylafax-defaults
  setOptions
  initRolodex
  initBatch
*** orig/tkhylafax.man	Wed Sep 24 13:47:46 1997
--- tkhylafax.man	Wed Sep 24 13:49:22 1997
***************
*** 235,240 ****
--- 235,242 ----
  A legal line in this file, for example, would be ``*faxResolution: high''. A 
  ``!'' in the first column of any line is treated as a comment. A 
  sample .tkhylafax-default file is included in this distribution.
+ If no such file in the user's home directory is found, the system-wide
+ resource file '/usr/local/lib/tkhylafax-3.0/tkhylafax-defaults is used.
  
  I'm thinking of adding a 
  .I few

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 26 Sep 1997 13:46:10 --100
From: coppelli@sun3.dsea.unipi.it (Alessandro Coppelli)
To: flexfax@sgi.com
Subject: flexfax: Problem with ps2fax
Sender: owner-flexfax@celestial.com

Hi to all.

  I have small problem when I send fax :
  
   if I send only text file all ok ( on server, but from client WIN95 WHFC
   no because the files must be *.ps )
   
   Then I prove this:
   
     on server I get two files poscript and run 
     
      ps2fax first_file.ps
      
      ps2fax second_file.ps
      
   The first all ok ( the file is poscript from Netscape --> Save As 
-->Poscript)
   
   The second is a cover sheet poscript file ( faxcover.ps ) from program tgif
   and then make_faxcover
   
   
   I get this error :
   
   Error:/undefined in to
   Operand stack:
   
   Execution stack:
   
    %interp_exit --nostringval-- nostringval--  nostringval--  false 
--nostringval--  --nostringval--
    Dictionary stack:
     --dict:603/631--   dict:0/20--  --dict:39/200--
     Current allocation mode is local
     Current file position is 5126
     
     
     
   I use Aladdin Ghoscript 3.33 on Linux 2.0.30
   
   
   Any idea for this error ?
   
   
   
   
   
   

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 30 Sep 1997 08:16:23 +0200
From: Ulrich Eckhardt <uli@transcom.de>
Organization: Transcom
To: Alessandro Coppelli <coppelli@sun3.dsea.unipi.it>
CC: flexfax@sgi.com
Subject: Re: flexfax: Problem with ps2fax
Sender: owner-flexfax@celestial.com

Alessandro Coppelli wrote:
> 
> Hi to all.
> 
>   I have small problem when I send fax :
> 
>    if I send only text file all ok ( on server, but from client WIN95 WHFC
>    no because the files must be *.ps )
> 
>    Then I prove this:

>    I use Aladdin Ghoscript 3.33 on Linux 2.0.30
> 
> 
>    Any idea for this error ?

Hi,

update to  Ghostscript 4.0 or 5.0 and all your problems
should gone (including your problems that whfc sends only
a cover sheet).

Uli
-- 
Ulrich Eckhardt		                   mailto:uli@transcom.de
http://people.frankfurt.netsurf.de/uli
Truly great madness can not be achieved without significant
intelligence. (Henrik Tikkanen)

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 30 Sep 1997 10:33:56 --100
From: coppelli@sun3.dsea.unipi.it (Alessandro Coppelli)
To: flexfax@sgi.com
Subject: flexfax: SUMMARY: Problem with gs ps2fax and whfc
Sender: owner-flexfax@celestial.com


 Hi to all.
 
 PROBLEM
 =======
 
 a) When I send a fax with WHFC only cover sheet is send
 
 b)  I received this error : 
 
 Sep 22 03:03:13 apps FaxSend[22790]: Client does not support document page
 width, max remote page width 1728 pixels, image width 1734 pixels
 
 RESOLUTION
 ==========
 
  a) In box Win95 with Office ( Editor Word )
   
     Add one printer ( APPLE LASER WRITER PLUS with make file on hard disk)
     
     Propriets --> PostScript --> EPS format 
     
     
  b) Change in ps2fax this line from:
       

        $CAT $fil | $PS -q \

    to:
    
        $CAT $fil | $SED -e 's/dup \/PageSize \[612 792\] put//g' | $PS -q \
        
        
   
   Thank you very much to  DAVE WRESKI
   
   Alessandro

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 30 Sep 1997 19:16:16 -0300 (GMT+3)
From: Ben Parker <ben@dha.unon.org>
To: Zdravko Balorda <zdravc@lie.fe.uni-lj.si>
cc: flexfax@sgi.com
Subject: Re: flexfax: error mail
Sender: owner-flexfax@celestial.com

Dirk's faxmailer stores the original as text and then returns it on
failure. You can then forward it back to the number to try again.

Ben


On Sat, 20 Sep 1997, Zdravko Balorda wrote:

> Hi,
> 
> on error hylafax returns a message to the sender
> along with coded fax. I wonder what one could do
> with this coded fax? Wouldn't it be better if the
> job got saved somehow in, say, an error directory.
> Hylafax mail could inform sender how to resend it.
> 
> Thanks, Zdravko.
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: sendpage+hfaxd, SNPP DATA broken? (4.0pl1 Sol2.5binary)
To: frohme@memphis.edu (Mike Frohme)
Date: Sun, 9 Nov 1997 08:59:11 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Mike Frohme wrote:
    
    Hello,
    
    	Just picked up HylaFax 4.0pl1 (Solaris 2.5, binary dist
    	from SGI) and have installed.  Fax features work great,
    	but I'm having minor problems w/Alpha paging:
    	
    	sendpage works great when you specify the message in the
    	command line.  Specifying the message via STDIN fails - the
    	read never completes.
    
    	It appears that the MESS command is AOK, but the DATA command
    	hangs in this distribution.  
    
    telnet localhost 444
    Trying 127.0.0.1...
    Connected to localhost.
    Escape character is '^]'.
    220 palantir.memphis.edu SNPP server (HylaFAX (tm) Version 4.0pl1)
    ready.
    login XX
    230 User XX logged in.
    page frohme
    250 Pager ID accepted; provider: 9,XXXXXXX pin: XXXXXXX jobid: 35.
    DATA
    354 Begin message input; end with <CRLF>'.'<CRLF>.
    Test message
    Line #2
[perhaps here was a line containing only a dot "." and the MTA
 cuts your mail here because the dot terminates SMTP also]

Please enable the serverTracing for hfaxd(1M) to see the
SNPP client/server protocol for this session, e.g. set

LogFacility:            local0          # syslog facility (want this 1st))
ServerTracing:          0x003           # server-op + c/s-prot

in lib/hfaxd.conf and re-run this test.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 10 Nov 1997 08:37:22 -0600
From: Mike Frohme <frohme@memphis.edu>
Organization: Information Systems - University of Memphis
To: flexfax@sgi.com
Subject: Re: flexfax: sendpage+hfaxd, SNPP DATA broken? (4.0pl1 Sol2.5binary)
Sender: owner-flexfax@celestial.com

> Matthias.Apitz@SOFTCON.de wrote:
>
> Mike Frohme wrote:
>
>     Hello,
>
>         Just picked up HylaFax 4.0pl1 (Solaris 2.5, binary dist
>         from SGI) and have installed.  Fax features work great,
>         but I'm having minor problems w/Alpha paging:
>
>         sendpage works great when you specify the message in the
>         command line.  Specifying the message via STDIN fails - the
>         read never completes.
>
>         It appears that the MESS command is AOK, but the DATA command
>         hangs in this distribution.
>
>     telnet localhost 444
>     Trying 127.0.0.1...
>     Connected to localhost.
>     Escape character is '^]'.
>     220 palantir.memphis.edu SNPP server (HylaFAX (tm) Version 4.0pl1)
>     ready.
>     login XX
>     230 User XX logged in.
>     page frohme
>     250 Pager ID accepted; provider: 9,XXXXXXX pin: XXXXXXX jobid: 35.
>     DATA
>     354 Begin message input; end with <CRLF>'.'<CRLF>.
>     Test message
>     Line #2
> [perhaps here was a line containing only a dot "." and the MTA
>  cuts your mail here because the dot terminates SMTP also]
>
        Yes, I manually entered a '.', but it doesn't work.  It *is*
        reading data on the port, because the server issues a
        550 Error, message too long; max 128 characters,  if I go
        over the limit.

        Its like a CRLF semantics problem or something, and the
        server isn't seeing it quite right.  Its obviously still
        reading, just not seeing a *strict* <CRLF>.<CRLF>

        I've tried the telnet from a PC, and programmed keys to send
        LF, CR, LFCR, CRLF, etc...  No combination of those seems
        able to get the DATA read to finish successfully.

> Please enable the serverTracing for hfaxd(1M) to see the
> SNPP client/server protocol for this session, e.g. set
>
> LogFacility:            local0          # syslog facility (want this 1st))
> ServerTracing:          0x003           # server-op + c/s-prot
>
> in lib/hfaxd.conf and re-run this test.
>
        Here's the log file for... sendpage -v -p frohme < textfile

Nov  9 17:31:03 palantir HylaFAX[18567]: LOGI frohme
Nov  9 17:31:03 palantir HylaFAX[18567]: SITE HELP NOTIFY
Nov  9 17:31:03 palantir HylaFAX[18567]: LEVE 1
Nov  9 17:31:03 palantir HylaFAX[18567]: ??? Mike Frohme
Nov  9 17:31:03 palantir HylaFAX[18567]: ??? 12
Nov  9 17:31:03 palantir HylaFAX[18567]: ??? 3
Nov  9 17:31:03 palantir HylaFAX[18567]: ??? frohme@palantir.memphis.edu
Nov  9 17:31:03 palantir HylaFAX[18567]: ??? none
Nov  9 17:31:03 palantir HylaFAX[18567]: ??? NO
Nov  9 17:31:03 palantir HylaFAX[18567]: PAGE frohme
Nov  9 17:31:03 palantir HylaFAX[18567]: Filesystem has SysV-style file creation
semantics.
Nov  9 17:31:03 palantir HylaFAX[18567]: DATA

        That's it.

-- 
Mike Frohme                                     frohme@memphis.edu
Information Systems - The University of Memphis       901/678-8820

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: sendpage+hfaxd, SNPP DATA broken? (4.0pl1 Sol2.5binary)
To: flexfax@sgi.com
Date: Mon, 10 Nov 1997 15:21:24 +0100 (MEZ)
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Mike Frohme wrote:
    
    Matthias.Apitz@SOFTCON.de wrote:
    > 
    > Mike Frohme wrote:
    > 
    >     Hello,
    > 
    >         Just picked up HylaFax 4.0pl1 (Solaris 2.5, binary dist
    >         from SGI) and have installed.  Fax features work great,
    >         but I'm having minor problems w/Alpha paging:
    > 
    >         sendpage works great when you specify the message in the
    >         command line.  Specifying the message via STDIN fails - the
    >         read never completes.
    > 
    >         It appears that the MESS command is AOK, but the DATA command
    >         hangs in this distribution.
    > 
    >     telnet localhost 444
    >     Trying 127.0.0.1...
    >     Connected to localhost.
    >     Escape character is '^]'.
    >     220 palantir.memphis.edu SNPP server (HylaFAX (tm) Version 4.0pl1)
    >     ready.
    >     login XX
    >     230 User XX logged in.
    >     page frohme
    >     250 Pager ID accepted; provider: 9,XXXXXXX pin: XXXXXXX jobid: 35.
    >     DATA
    >     354 Begin message input; end with <CRLF>'.'<CRLF>.
    >     Test message
    >     Line #2
    > [perhaps here was a line containing only a dot "." and the MTA
    >  cuts your mail here because the dot terminates SMTP also]
    > 
    	Yes, I manually entered a '.', but it doesn't work.  It *is*
    	reading data on the port, because the server issues a 
    	550 Error, message too long; max 128 characters,  if I go
    	over the limit.

Oh. This is another situation. The sendpage(1) failed
because the max length of the message is reached (and the DATA
command is aborted and hfaxd(1M) isn't waiting for a "." anymore).
Just say "quit" and hfaxd(1M) will terminate this session.
Does it also ignores the "<CRLF>.<CRLF>" if you don't reach the limit?

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 10 Nov 1997 09:05:10 -0600
From: Mike Frohme <frohme@memphis.edu>
Organization: Information Systems - University of Memphis
To: flexfax@sgi.com
Subject: Re: flexfax: sendpage+hfaxd, SNPP DATA broken? (4.0pl1 Sol2.5binary)
Sender: owner-flexfax@celestial.com

Matthias.Apitz@SOFTCON.de wrote:
> Mike Frohme wrote:
>     Matthias.Apitz@SOFTCON.de wrote:
>     > Mike Frohme wrote:
>     >
>     >     Hello,
>     >
>     >         Just picked up HylaFax 4.0pl1 (Solaris 2.5, binary dist
>     >         from SGI) and have installed.  Fax features work great,
>     >         but I'm having minor problems w/Alpha paging:
>     >
>     >         sendpage works great when you specify the message in the
>     >         command line.  Specifying the message via STDIN fails - the
>     >         read never completes.
>     >
>     >         It appears that the MESS command is AOK, but the DATA command
>     >         hangs in this distribution.
>     >
>     >     telnet localhost 444
>     >     Trying 127.0.0.1...
>     >     Connected to localhost.
>     >     Escape character is '^]'.
>     >     220 palantir.memphis.edu SNPP server (HylaFAX (tm) Version 4.0pl1)
>     >     ready.
>     >     login XX
>     >     230 User XX logged in.
>     >     page frohme
>     >     250 Pager ID accepted; provider: 9,XXXXXXX pin: XXXXXXX jobid: 35.
>     >     DATA
>     >     354 Begin message input; end with <CRLF>'.'<CRLF>.
>     >     Test message
>     >     Line #2
>     > [perhaps here was a line containing only a dot "." and the MTA
>     >  cuts your mail here because the dot terminates SMTP also]
>     >
>         Yes, I manually entered a '.', but it doesn't work.  It *is*
>         reading data on the port, because the server issues a
>         550 Error, message too long; max 128 characters,  if I go
>         over the limit.
> 
> Oh. This is another situation. The sendpage(1) failed
> because the max length of the message is reached (and the DATA
> command is aborted and hfaxd(1M) isn't waiting for a "." anymore).
> Just say "quit" and hfaxd(1M) will terminate this session.
> Does it also ignores the "<CRLF>.<CRLF>" if you don't reach the limit?
> 
        Yes, that's the point.  It *never* (as far as I can tell) sees
        the "<CRLF>.<CRLF>" in DATA mode.

        I added the comments about "550 error" only to validate that the 
        server is actually reading data from the channel.  Its just not
        seeing the end-of-input.

-- 
Mike Frohme                                     frohme@memphis.edu
Information Systems - The University of Memphis       901/678-8820
-- 
Mike Frohme                                     frohme@memphis.edu
Information Systems - The University of Memphis       901/678-8820

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
X-Authentication-Warning: boxhill.com: Host localhost.boxhill.com didn't use HELO protocol
To: Mike Frohme <frohme@memphis.edu>
cc: flexfax@sgi.com
Subject: Re: flexfax: sendpage+hfaxd, SNPP DATA broken? (4.0pl1 Sol2.5binary) 
Date: Tue, 11 Nov 1997 10:48:39 -0500
From: "Thomas A. Szybist" <szybist@boxhill.com>
Sender: owner-flexfax@celestial.com

In message <346722A6.32F389DC@memphis.edu>, Mike Frohme writes:
> Matthias.Apitz@SOFTCON.de wrote:
> > Mike Frohme wrote:
> >     Matthias.Apitz@SOFTCON.de wrote:
> >     > Mike Frohme wrote:
> >     >
> >     >     Hello,
> >     >
> >     >         Just picked up HylaFax 4.0pl1 (Solaris 2.5, binary dist
> >     >         from SGI) and have installed.  Fax features work great,
> >     >         but I'm having minor problems w/Alpha paging:
> >     >
> >     >         sendpage works great when you specify the message in the
> >     >         command line.  Specifying the message via STDIN fails - the
> >     >         read never completes.
> >     >
> >     >         It appears that the MESS command is AOK, but the DATA command
> >     >         hangs in this distribution.
> >     >
> >     >     telnet localhost 444
> >     >     Trying 127.0.0.1...
> >     >     Connected to localhost.
> >     >     Escape character is '^]'.
> >     >     220 palantir.memphis.edu SNPP server (HylaFAX (tm) Version 4.0pl1)
> >     >     ready.
> >     >     login XX
> >     >     230 User XX logged in.
> >     >     page frohme
> >     >     250 Pager ID accepted; provider: 9,XXXXXXX pin: XXXXXXX jobid: 35.
> >     >     DATA
> >     >     354 Begin message input; end with <CRLF>'.'<CRLF>.
> >     >     Test message
> >     >     Line #2
> >     > [perhaps here was a line containing only a dot "." and the MTA
> >     >  cuts your mail here because the dot terminates SMTP also]
> >     >
> >         Yes, I manually entered a '.', but it doesn't work.  It *is*
> >         reading data on the port, because the server issues a
> >         550 Error, message too long; max 128 characters,  if I go
> >         over the limit.
> > 
> > Oh. This is another situation. The sendpage(1) failed
> > because the max length of the message is reached (and the DATA
> > command is aborted and hfaxd(1M) isn't waiting for a "." anymore).
> > Just say "quit" and hfaxd(1M) will terminate this session.
> > Does it also ignores the "<CRLF>.<CRLF>" if you don't reach the limit?
> > 
>         Yes, that's the point.  It *never* (as far as I can tell) sees
>         the "<CRLF>.<CRLF>" in DATA mode.
> 
>         I added the comments about "550 error" only to validate that the 
>         server is actually reading data from the channel.  Its just not
>         seeing the end-of-input.
> 
> -- 
> Mike Frohme                                     frohme@memphis.edu
> Information Systems - The University of Memphis       901/678-8820
> -- 
> Mike Frohme                                     frohme@memphis.edu
> Information Systems - The University of Memphis       901/678-8820

I tried to reproduce this, but was unable to do so.  Unfortunately, 
I have upgraded my systems to Solaris 2.6 and recompiled everything.
If you are interested in trying this binary, let me know.

I will try to run this with the original binary on Solaris 2.5, but
it may be a little while.  

Tom Szybist
szybist@boxhill.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
X-Authentication-Warning: boxhill.com: Host localhost.boxhill.com didn't use HELO protocol
To: Mike Frohme <frohme@memphis.edu>
cc: flexfax@sgi.com
Subject: Re: flexfax: sendpage+hfaxd, SNPP DATA broken? (4.0pl1 Sol2.5binary) 
Date: Tue, 11 Nov 1997 11:58:47 -0500
From: "Thomas A. Szybist" <szybist@boxhill.com>
Sender: owner-flexfax@celestial.com

In message <346722A6.32F389DC@memphis.edu>, Mike Frohme writes:
> Matthias.Apitz@SOFTCON.de wrote:
> > Mike Frohme wrote:
> >     Matthias.Apitz@SOFTCON.de wrote:
> >     > Mike Frohme wrote:
> >     >
> >     >     Hello,
> >     >
> >     >         Just picked up HylaFax 4.0pl1 (Solaris 2.5, binary dist
> >     >         from SGI) and have installed.  Fax features work great,
> >     >         but I'm having minor problems w/Alpha paging:
> >     >
> >     >         sendpage works great when you specify the message in the
> >     >         command line.  Specifying the message via STDIN fails - the
> >     >         read never completes.
> >     >
> >     >         It appears that the MESS command is AOK, but the DATA command
> >     >         hangs in this distribution.
> >     >
> >     >     telnet localhost 444
> >     >     Trying 127.0.0.1...
> >     >     Connected to localhost.
> >     >     Escape character is '^]'.
> >     >     220 palantir.memphis.edu SNPP server (HylaFAX (tm) Version 4.0pl1)
> >     >     ready.
> >     >     login XX
> >     >     230 User XX logged in.
> >     >     page frohme
> >     >     250 Pager ID accepted; provider: 9,XXXXXXX pin: XXXXXXX jobid: 35.
> >     >     DATA
> >     >     354 Begin message input; end with <CRLF>'.'<CRLF>.
> >     >     Test message
> >     >     Line #2
> >     > [perhaps here was a line containing only a dot "." and the MTA
> >     >  cuts your mail here because the dot terminates SMTP also]
> >     >
> >         Yes, I manually entered a '.', but it doesn't work.  It *is*
> >         reading data on the port, because the server issues a
> >         550 Error, message too long; max 128 characters,  if I go
> >         over the limit.
> > 
> > Oh. This is another situation. The sendpage(1) failed
> > because the max length of the message is reached (and the DATA
> > command is aborted and hfaxd(1M) isn't waiting for a "." anymore).
> > Just say "quit" and hfaxd(1M) will terminate this session.
> > Does it also ignores the "<CRLF>.<CRLF>" if you don't reach the limit?
> > 
>         Yes, that's the point.  It *never* (as far as I can tell) sees
>         the "<CRLF>.<CRLF>" in DATA mode.
> 
>         I added the comments about "550 error" only to validate that the 
>         server is actually reading data from the channel.  Its just not
>         seeing the end-of-input.
> 
> -- 
> Mike Frohme                                     frohme@memphis.edu
> Information Systems - The University of Memphis       901/678-8820
> -- 
> Mike Frohme                                     frohme@memphis.edu
> Information Systems - The University of Memphis       901/678-8820

Hi,

I did manage to get a Solaris 2.5 machine that even had 
the binary distribution on it.  I was able to reproduce
the problem on that machine.  Truss on hfaxd showed a
read attempt after the ".".  Hmmm...  Then it justed polled.
Clearly it was missing the ".".  

I replaced hfaxd with what I compiled for Solaris 2.6 and 
that worked fine. The system call sequence looked pretty
much the same for both binaries.

All I can think of is that I compiled the 2.5 binaries with
gcc 2.7.2 and the 2.6 binary with 2.7.2.3.  I have heard
of some optimizer problems with early versions of 2.7.2.

Again, you can contact me if you would like to try this binary.

Tom Szybist
szybist@boxhill.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 12 Nov 1997 12:19:58 -0600
From: Mike Frohme <frohme@memphis.edu>
Organization: Information Systems - University of Memphis
To: "Thomas A. Szybist" <szybist@boxhill.com>
CC: flexfax@sgi.com
Subject: Re: flexfax: sendpage+hfaxd, SNPP DATA broken? (4.0pl1 Sol2.5binary)
Sender: owner-flexfax@celestial.com

Thomas A. Szybist wrote:
> 
> In message <3469C29E.461572DB@memphis.edu>, Mike Frohme writes:
> > Thomas A. Szybist wrote:
> > >
> > > Did you get my reply to this?  Just in case you didn't, you can pick up hfaxd
> > > from my 2.6 binary from ftp://24.3.148.6.   The complete binary
> > > is there as well.  You should be able to just replace hfaxd. (Be
> > > certain to make it executable.)
> > >
> > > This machine may be unavailable overnight.  But it should be up
> > > tomorrow morning. (Wednesday, 11/12 ).
> > >
> >       Thanks.  I was out yesterday afternoon, and will be this one as well.
> >       I picked up both the exec and the dist.
> >
> > > Let me know how you make out.
> > >
> > > Tom Szybist
> > > szybist@boxhill.com
> >
> >       I seem to have buggered something up.  I stopped the server, etc...
> >       moved the binary, dropped in the the new one, made sure the file
> >       ownerships/permissions/setuid were correct, and so forth.
> >
> >       It accepts things AOK - the DATA problem seems fixed, but it
> >       won't get 'round to talking to the modem.  Seems the queueing
> >       system has a problem.  On sendpage, etc... I often (not always)
> >       get console errors of the form:
> >
> > Nov 12 08:34:54 palantir FaxQueuer[25191]: HylaClient::send: client/24919: Cannot open
> > FIFO: No such device or address
> >
> >       As far as I can tell, its there.  When I run one of the programs
> >       under truss control, I get:
> >
> > open("FIFO", O_WRONLY|O_NDELAY)                 Err#6 ENXIO
> >
> >       then the no such device or address error.  Naturally, I never
> >       get beyond "waiting for modem to come ready" status.
> >
> >       Funny thing is, when I go back to the original version of hfaxd,
> >       I still get the same errors.  Hmmm...  ideas?
> >
> > --
> > Mike Frohme                                     frohme@memphis.edu
> > Information Systems - The University of Memphis       901/678-8820
> 
> Here is what my FIFO file looks like:
> 
> orff rszybist(7)> ls -l FIFO
> prw-------   1 uucp     uucp           0 Nov 12 11:37 FIFO|
> 
> It should be in /var/spool/fax
> 
> What does faxstat say?
> 
> You might want to stop faxq, remove the FIFO file,
> and restart.   faxq should create the FIFO if
> it doesn't exist.

	I did a variety of things.  My FIFO looked just like yours.  I'd
	stopped and restarted things to no avail.  Re-added the modem via
	faxaddmodem, and so forth - still no luck.

	Turns out that its fixable by re-running faxsetup. This also gets
	the new binary to work correctly. Just drop in the binary and
	run faxsetup - it works fine. I guess I was too out-of-it this
	AM to think of that.

	Looks like the DATA problem is resolved, as you might expect.
	Thanks for the help, the new version, and for your suggestions.

-- 
Mike Frohme                                     frohme@memphis.edu
Information Systems - The University of Memphis       901/678-8820

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: "Herman Harless" <herman@citynet.net>
To: <flexfax@sgi.com>
Subject: flexfax: Weird faxcover problem (not FAQ #36)
Date: Wed, 24 Sep 1997 11:58:00 -0400
Sender: owner-flexfax@celestial.com

I read FAQ #36 & generated a fax cover sheet for hylafax.  The sheet itself
works fine, and the variables generally go in their places.  However, I
have one of those "strange" problems.

I'm using the same font throughout the the form.

The template prints fine, everything's in its place

In the variable fields, certain but not all, capitol letters are dropped. 
I've looked at the ps document, and the data is getting to it, but not
comming out on the fax.  Ghostscript doesn't show the letters either. 
These same letters print fine in the template.

I'm using gs ver 5.03
              hylafax ver 4.0pl1

any ideas?

thanx
herman
herman@citynet.net

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Raed Nashef <nash_ra@eng.printronix.com>
Date: Wed, 24 Sep 1997 12:31:56 -0700 (PDT)
Subject: flexfax: Can not start HylaFAX server processes
To: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

------------- Begin Forwarded Message -------------

>From nash_ra Wed Sep 24 12:28:41 1997
To: sam@engr.sgi.com
Subject: Can not start HylaFAX server processes
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-MD5: yHU78oEN6uI89i4aRkQiqA==

Hello Sam,

I hope it is OK to seak your help troubleshooting this problem.
I installed HylaFAX on a SUN Ulta Sparc machine. Every thing looked
OK until I ran the faxsetup script. It is giving me the following error
when trying to start the server processes faxq:
"ld.so.1: /opt/hylafax/4.0/spool/fax/bin/faxq: fatal: relocation error: symbol 
not found: TIFFFaxBlackTable: referenced in /opt/hylafax/4.0/spool/fax/bin/faxq"

Any ideas?

REM: I installed the latest tiff software before installing HylaFAX.

Thanks,
Raed Nashef
nash_ra@eng.printronix.com
System Specialist
PRINTRONIX
------------- End Forwarded Message -------------

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 25 Sep 1997 16:01:53 -0700 (PDT)
From: RHS Linux User <faxmail@page.esgroup.net>
To: flexfax@sgi.com
Subject: flexfax: too many unknown responces to message block transmit
Sender: owner-flexfax@celestial.com

Hi.  I'm getting an error on almost every page that I send out.
Basically, it says:

Protocol Failure: too manyh unknown responces to message block transmit.

Wierd thing with this is that it still goes out most of the time.  Quite
often, it will send out the page, and return this error, but the page
still goes through.  Other times, it gets the error, but doesn't send the
page.  Still other times, it works, with no error.

Does anyone have any suggestions?

Mike Oswell
ESI Communications

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: "Alan Sparks" <asparks@nss.harris.com>
To: <flexfax@sgi.com>
Subject: flexfax: WHFC and NT - '': Command not understood - progress?
Date: Fri, 26 Sep 1997 17:36:10 -0700
Sender: owner-flexfax@celestial.com

I've noted in the mailing list that several people have reported having
problems getting WHFC to work on NT4.0, getting a '': Command not understood
or such error message.  Well, I guess I'm on that list too.

I've tried telnet'ing into the Sun (SunOS 4.1.4) server that is running
hfaxd.  Telnets from UNIX boxes (HP, Sun) work OK, but every command from
the NT telnet session gives the error above.

Noted mention of a debugging version out, haven't tried that yet...

Has anyone figured out what the problem might be, yet?  Thanks in advance
for any updates.

-Alan

---------
Alan Sparks, IS Engineering Support                   <asparks@harris.com>
Harris Network Support Systems, Camarillo CA    (805) 389-2430

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Sat, 27 Sep 1997 09:40:21 -0700
From: Bill Campbell <bill@dagney.celestial.com>
To: Mail Administrator <Postmaster@bdg.centrin.net.id>
Cc: flexfax@celestial.com, postmaster@jupiter.centrin.net.id
Subject: flexfax: Re: Mail System Error - Returned Mail
Reply-To: bill@celestial.com
Sender: owner-flexfax@celestial.com

I have been getting many of these messages which don't have sufficient
information to track back to the original address (i.e. there are NO
Received headers in the message you bounce).

We don't have anybody on our mailing list with an address that looks
remotely like this so I have to assume that it's coming from somebody elses
list exploder (somebody who doesn't have a clue how to set up a mailing
list properly so that their local propagation errors come back to them
instead of the original list owner).

On Sat, Sep 27, 1997 at 11:11:11AM +0700, Mail Administrator wrote:
>This Message was undeliverable due to the following reason:
>
>The following destination addresses were unknown (please check
>the addresses and re-mail the message):
>
>SMTP <ih@bdg.centrin.net.id>
>
>Please reply to Postmaster@bdg.centrin.net.id
>if you feel this message to be in error.

>Date: Fri, 26 Sep 1997 07:14:16 -0500
>From: Selim Jose Miselem <selim@opensys.hn>
>To: flexfax@sgi.com
>Subject: flexfax: Fax Sending stops after using whfc - PLEASE HELP!!!!
>
....body of message deleted....


-- 
Bill
--
INTERNET:   bill@Celestial.COM  Bill Campbell; Celestial Systems, Inc.
UUCP:               camco!bill  PO Box 820; 2835 82nd Avenue S.E. S-100
FAX:            (206) 232-9186  Mercer Island, WA 98040-0820; (206) 236-1676
URL: http://www.celestial.com/

Whenever is found what is called a paternal government, there is found 
state education. It has been discovered that the best way to insure 
implicit obedience is to commence tyranny in the nursery. 
    (1874) Benjamin Disraeli

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: sendpage and apache on Linux
To: aschlies@citynet.net (Tony Schliesser)
Date: Sat, 27 Sep 1997 13:18:35 +0100 (BST)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> 
> I am working on a simple sendpage/web gateway.  From the command line, a
> simple C program such as:
> 
>    execl("/usr/local/bin/sendpage","sendpage.web", "-q",
> "-ptony","hello...testing",NULL);
> }

I don't know why you are using C; a simple shell wrapper should work,
something like:

    #! /bin/sh
    echo "Status: 200 OK"            # strictly should be pending not OK code
                                     # but I don't have the HTTP spec to hand
    echo "Content-Type: text/html"
    echo
    echo "<HTML><HEAD><TITLE>Tony's Page Request Acceptance Page</TITLE>"
    echo "</HEAD><BODY><H1>Page Accepted</H1>Your page request has be queued"
    echo "for transmission.</BODY></HTML>"
    exec /usr/local/bin/sendpage -q -ptony "hello...testing" > /dev/null
               # sendpage command not verified
    
You could use text/plain and return the actual response from sendpage. To
include it in HTML, you will need to bracket it in <PRE></PRE> and  translate
& < > and possibly " into entities (e.g. use sed).

You might want to look at the cgiparse utility that comes with the CERN 
server.

> However, if I put the execl line into a C program and run as a cgi on a web
> server, it bombs with a server error.  

A valid CGI program must output at least "Location: url" or "Content-Type:
xxx/yyy" followed by a blank line.  "Status: 204 No output" ought also
to be OK, again followed by a blank line.

> 
> I can only assume that perhaps the output of sendpage (lots of texts) is
> the problem since the return value exec* is expecting is an int.  I don't

The output from sendpage is on standard output, or possibly standard error.
It ALSO returns an int return code.

Note that the above is telling you the requirements for a CGI script - it
is possible that sendpage has other requirements on its environment.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: "Kai O'Yang" <oyang@phoebe.fcit.monash.edu.au>
Subject: flexfax: Re: SUMMARY: What's "REMOTE HANGUP: DTE to DCE data underflow (code 43)"?
To: oyang@phoebe.fcit (Kai O'Yang)
Date: Mon, 29 Sep 1997 16:36:27 +1000 (EST)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

Hi again,

My original question:
>I'm trying to install a Hylafax server on a Solaris 2.5.1 x86 box with
>Hayes Accura 288 faxmodem. 
 
>Sometimes sendfax works but a lot of times it dies with:
>...
>Sep 23 13:16:17.05: [15317]: <-- data [924]
>Sep 23 13:16:17.05: [15317]: SENT 25500 bytes of data
>Sep 23 13:16:17.05: [15317]: <-- data [2]
>Sep 23 13:16:18.37: [15317]: SEND end page
>Sep 23 13:16:18.62: [15317]: --> [8:+FHNG:43]
>Sep 23 13:16:18.62: [15317]: REMOTE HANGUP: DTE to DCE data underflow (code 43)
>Sep 23 13:16:19.24: [15317]: --> [2:OK]
>Sep 23 13:16:19.24: [15317]: <-- [5:ATH0\r]
>Sep 23 13:16:19.37: [15317]: --> [2:OK]
>Sep 23 13:16:19.38: [15317]: SESSION END
> 
>Does anyone know what is the problem and how to fix it?

The problem was traced back to a faulty DB25<->DB9 adaptor which does not
have RTS/CTS lines wired! (stupid me. shouldn't have used those kind of
el cheapo cables.) After replacing the cable, setting up using 19200bps with
rtscts flow control, everything is fine.

Special thanks to David Woolley for helping me out.

Kai

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 29 Sep 1997 20:02:37 +0300 (EET DST)
From: Salih Tilkici <Salih.Tilkici@dut.net.tr>
Reply-To: Salih Tilkici <Salih.Tilkici@dut.net.tr>
To: Ben Parker <ben@dha.unon.org>
cc: David Woolley <david@djwhome.demon.co.uk>, flexfax@sgi.com
Subject: flexfax: Re: faxmail :attachment file ?
Sender: owner-flexfax@celestial.com

Hi,

I am working on faxmail mime converter script. I'm trying to convert tiff
to ps. I know that my converter script is working. Because I'm redirecting
a copy to another file and it seems ok. 

But when I try to display the doc file only From and subject displayed.
The attachment parts are displayed as blank. When I look at the source of
docq file, I can see the converted tiff part in postscript format. I think
it is merging two postscript files in wrong format.

How can solve this problem. Can anyone send an example (working;) 
converter script. I'm using a script like this: 
===========================================================
#!/bin/sh
   file=$1
   TMP=/tmp/debugfile
#This is for faxmail
   convert tiff:$file ps:-
#this is for testing
   convert tiff:$file ps:$TMP
===========================================================

Regards,

Salih Tilkici

PS: I'm using ImageMagick's convert program.

On Wed, 23 Jul 1997, Ben Parker wrote:

> 
> 
> On Tue, 22 Jul 1997, David Woolley wrote:
> 
> 
> > Looking at the source code for faxmail.c++ (took about 5 minutes) it is
> > expecting:
> > 
> > giftopnm $1 | pnmtops
> 
> Great. That helps a lot. I had tried %1 but not $1!
> 
> > It's running
> > 
> >   script GIF-file
> > 
> > where GIF-file is a temporary file.
> 
> Yes - now I can get some ps out the other end. Still need to work on it
> though....

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 30 Sep 1997 17:49:18 -0300 (GMT+3)
From: Ben Parker <ben@dha.unon.org>
To: Salih Tilkici <Salih.Tilkici@dut.net.tr>
cc: David Woolley <david@djwhome.demon.co.uk>, flexfax@sgi.com
Subject: flexfax: Re: faxmail :attachment file ?
Sender: owner-flexfax@celestial.com


This is what I now use for GIF. Haven't had time to try any other formats
yet (JPEG would be nice). Two scripts. The second, the perl script, hacks
the output to make it fit on one separate page almost full size (18/20). 

#/usr/local/sbin/faxmail/image/gif
#!/bin/bash
giftopnm -v $1 | ppmtopgm | pnmtops -rle -nocenter | /usr/local/sbin/faxmail/image/pnmpsfix.pl

#/usr/local/sbin/faxmail/image/pnmpsfix.pl
#!/usr/bin/perl
while (<STDIN>) {
	if ($_ =~ "scale") {
		($x,$y,$junk) = split(/\s/);
		$newx = $x * 18;
		$newy = $y * 18;
		print "showpage\n";
		print "15 25 translate\n";
	}
	print $_;
}



Good luck,

Ben


On Mon, 29 Sep 1997, Salih Tilkici wrote:

> Hi,
> 
> I am working on faxmail mime converter script. I'm trying to convert tiff
> to ps. I know that my converter script is working. Because I'm redirecting
> a copy to another file and it seems ok. 
> 
> But when I try to display the doc file only From and subject displayed.
> The attachment parts are displayed as blank. When I look at the source of
> docq file, I can see the converted tiff part in postscript format. I think
> it is merging two postscript files in wrong format.
> 
> How can solve this problem. Can anyone send an example (working;) 
> converter script. I'm using a script like this: 
> ===========================================================
> #!/bin/sh
>    file=$1
>    TMP=/tmp/debugfile
> #This is for faxmail
>    convert tiff:$file ps:-
> #this is for testing
>    convert tiff:$file ps:$TMP
> ===========================================================
> 
> Regards,
> 
> Salih Tilkici
> 
> PS: I'm using ImageMagick's convert program.
> 
> On Wed, 23 Jul 1997, Ben Parker wrote:
> 
> > 
> > 
> > On Tue, 22 Jul 1997, David Woolley wrote:
> > 
> > 
> > > Looking at the source code for faxmail.c++ (took about 5 minutes) it is
> > > expecting:
> > > 
> > > giftopnm $1 | pnmtops
> > 
> > Great. That helps a lot. I had tried %1 but not $1!
> > 
> > > It's running
> > > 
> > >   script GIF-file
> > > 
> > > where GIF-file is a temporary file.
> > 
> > Yes - now I can get some ps out the other end. Still need to work on it
> > though....
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 29 Sep 1997 10:47:55 -0700
From: Greg <greg@pocketscience.com>
To: flexfax@sgi.com
Subject: flexfax: Some Hylafax SNPP patches 
Sender: owner-flexfax@celestial.com

Hi,
    I've recently started using hylafax for outbound paging.  (IXO/TAP).
It is well suited for a high volume environment we plan to use it in, as
it supports queueing and multiple modems. Anyway, in my travels I have
encountered some problems and include my solutions. The first problem I
ran into was that I could not specify the TAP dialup number when sending
a page request. Hylafax supports a regular expression maching and
substitution on the PIN, but not the Number. The first section of the
patch enables the use of things such as  111111@18005551212 to be sent
to pin number 111111 at TAP dialup 18005551212 (using the rule
"(.*)\@(.*) \2/\1"). The second problem I encountered was the use of
this software on a solaris Autoclient system. Solaris supports what it
calls an autoclient, which is basicly a diskless workstation which uses
its local drive as a cache. I suspect the problem I encountered was
possibly due to this, perhaps a file updating on nfs and not in cache,
or vice versa. The symtoms are that faxes work properly, however when
sending a page you get an error message (if you use the -v -d flags in
sendpage) "scheduler NAK'd request", furthermore, your syslog will have
messages from FaxQueuer stating:"null or missing number in job request".
The second patch included seems to aleviate this problem by updating the
job request   after the tap number and pin have been filled into the job
request. These patches were made on 4.0-pl1, on a Solaris-x86 2.5.1
machine, using GCC as the compiler. The patches are to be applied to
SNPPServer.c++, located in the hfaxd source subdirectory. I hope you
find this information usefull..

- Greg


*** SNPPServer.c++      Tue Nov 26 15:21:51 1996
--- SNPPServer.c++.new  Fri Sep 26 15:18:44 1997
***************
*** 300,305 ****
--- 300,306 ----
                            pin = cp;
                            if (re)                     // do
substitutions
                                subRHS(pin, *re, pagerID);
+                               subRHS(number, *re, pagerID);
                        } else {                        // <dialstring>
                            number = np;
                            pin = pagerID;
***************
*** 1118,1123 ****
--- 1119,1125 ----
            , (const char*) curJob->jobid
        );
        msgs.append(curJob->jobid);
+         updateJobOnDisk(*curJob, emsg);
      } else
        reply(554, "%s.", (const char*) emsg);
      initSNPPJob();                            // reset job-related
state

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Some Hylafax SNPP patches
To: greg@pocketscience.com (Greg)
Date: Tue, 30 Sep 1997 14:05:39 +0200 (MSZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Greg wrote:
    
    Hi,
        I've recently started using hylafax for outbound paging.  (IXO/TAP).
    It is well suited for a high volume environment we plan to use it in, as
    it supports queueing and multiple modems. Anyway, in my travels I have
    encountered some problems and include my solutions. The first problem I
    ran into was that I could not specify the TAP dialup number when sending
    a page request. Hylafax supports a regular expression maching and
    substitution on the PIN, but not the Number. The first section of the
    patch enables the use of things such as  111111@18005551212 to be sent
    to pin number 111111 at TAP dialup 18005551212 (using the rule
    "(.*)\@(.*) \2/\1").

The pager identifier specified with the "-p pid" in sendpage(1)
is send via SNPP in the PAGEr directive to the SNPP-server (e.g.
to hfaxd(1M)). The PAGEr directive in RFC 1861 has no room for
a specification of the IXO/TAP service provider. That's why the
hfaxd(1M) uses a file pagermap(4F) to map the pager identifier to
a service provider phone number. IMHO this mapping should be
a privilege part of the server daemon and not in the hand of
a user also.

                            The second problem I encountered was the use of
    this software on a solaris Autoclient system. Solaris supports what it
    calls an autoclient, which is basicly a diskless workstation which uses
    its local drive as a cache. I suspect the problem I encountered was
    possibly due to this, perhaps a file updating on nfs and not in cache,
    or vice versa. The symtoms are that faxes work properly, however when
    sending a page you get an error message (if you use the -v -d flags in
    sendpage) "scheduler NAK'd request", furthermore, your syslog will have
    messages from FaxQueuer stating:"null or missing number in job request".
    The second patch included seems to aleviate this problem by updating the
    job request   after the tap number and pin have been filled into the job
    request. These patches were made on 4.0-pl1, on a Solaris-x86 2.5.1
    machine, using GCC as the compiler. The patches are to be applied to
    SNPPServer.c++, located in the hfaxd source subdirectory. I hope you
    find this information usefull..

I'll put this on my list of bugs but I'm not sure about this
solution (see also the comment in SNPPServer.c++ at line 1094).

Thanks anyway.

	matthias
    
    *** SNPPServer.c++      Tue Nov 26 15:21:51 1996
    --- SNPPServer.c++.new  Fri Sep 26 15:18:44 1997
    ***************
    *** 300,305 ****
    --- 300,306 ----
                                pin = cp;
                                if (re)                     // do
    substitutions
                                    subRHS(pin, *re, pagerID);
    +                               subRHS(number, *re, pagerID);
                            } else {                        // <dialstring>
                                number = np;
                                pin = pagerID;
    ***************
    *** 1118,1123 ****
    --- 1119,1125 ----
                , (const char*) curJob->jobid
            );
            msgs.append(curJob->jobid);
    +         updateJobOnDisk(*curJob, emsg);
          } else
            reply(554, "%s.", (const char*) emsg);
          initSNPPJob();                            // reset job-related
    state

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 30 Sep 1997 10:52:14 +0200
From: Gregor Longariva <Gregor.Longariva@informatik.uni-erlangen.de>
Reply-To: Gregor Longariva <grlongar@cip.informatik.uni-erlangen.de>
To: flexfax@sgi.com
Subject: flexfax: Problems receiving faxes with USR Sportser
Sender: owner-flexfax@celestial.com

Hi!

Sometimes I have problems to receive faxes with my USR Sportser
33.6 voice, running on a Linux server.
My Problem is, that the modem (or Hylafax?) refuses faxes which
are not received 100% correctly.

Can anybody help me?

Thanx!

Here's the log:

----------------------------------------------------------------

An attempt to receive facsimile on fax failed because:

    T.30 T2 timeout, expected page not received

    ---- Transcript of session follows ----

Sep 30 10:24:28.86: [  227]: SESSION BEGIN 00000257 499131640572
Sep 30 10:24:29.02: [  227]: <-- [4:ATA\r]
Sep 30 10:24:31.37: [  227]: --> [3:FAX]
Sep 30 10:24:31.45: [  227]: ANSWER: FAX CONNECTION
Sep 30 10:24:31.50: [  227]: RECV FAX: begin
Sep 30 10:24:33.93: [  227]: --> [7:CONNECT]
Sep 30 10:24:35.32: [  227]: --> [2:OK]
Sep 30 10:24:35.32: [  227]: <-- [9:AT+FRH=3\r]
Sep 30 10:24:38.42: [  227]: --> [0:]
Sep 30 10:24:38.59: [  227]: MODEM <Empty line>
Sep 30 10:24:38.60: [  227]: --> [2:OK]
Sep 30 10:24:38.60: [  227]: DELAY 1500 ms
Sep 30 10:24:40.11: [  227]: <-- [9:AT+FTH=3\r]
Sep 30 10:24:40.14: [  227]: --> [7:CONNECT]
Sep 30 10:24:40.16: [  227]: --> [7:CONNECT]
Sep 30 10:24:42.23: [  227]: --> [2:OK]
Sep 30 10:24:42.23: [  227]: <-- [9:AT+FRH=3\r]
Sep 30 10:24:42.69: [  227]: --> [7:CONNECT]
Sep 30 10:24:44.14: [  227]: --> [2:OK]
Sep 30 10:24:44.14: [  227]: REMOTE TSI "+49 9132 40599"
Sep 30 10:24:44.14: [  227]: <-- [9:AT+FRH=3\r]
Sep 30 10:24:44.15: [  227]: --> [7:CONNECT]
Sep 30 10:24:44.48: [  227]: --> [2:OK]
Sep 30 10:24:44.48: [  227]: REMOTE wants 9600 bit/s
Sep 30 10:24:44.48: [  227]: REMOTE wants page width 1728 pixels in 215 mm
Sep 30 10:24:44.48: [  227]: REMOTE wants unlimited page length
Sep 30 10:24:44.48: [  227]: REMOTE wants 3.85 line/mm
Sep 30 10:24:44.48: [  227]: REMOTE wants 2-D MR
Sep 30 10:24:44.48: [  227]: RECV training at v.29 9600 bit/s
Sep 30 10:24:44.48: [  227]: <-- [10:AT+FRM=96\r]
Sep 30 10:24:44.82: [  227]: --> [7:CONNECT]
Sep 30 10:24:46.50: [  227]: RECV: TCF 2011 bytes, 5% non-zero, 1900
zero-run
Sep 30 10:24:46.50: [  227]: --> [10:NO CARRIER]
Sep 30 10:24:46.50: [  227]: DELAY 75 ms
Sep 30 10:24:46.59: [  227]: <-- [9:AT+FTH=3\r]
Sep 30 10:24:46.78: [  227]: --> [7:CONNECT]
Sep 30 10:24:47.96: [  227]: --> [2:OK]
Sep 30 10:24:47.96: [  227]: TRAINING succeeded
Sep 30 10:24:47.96: [  227]: <-- [10:AT+FRM=96\r]
Sep 30 10:24:48.51: [  227]: --> [7:CONNECT]
Sep 30 10:24:48.51: [  227]: RECV: begin page
Sep 30 10:26:29.33: [  227]: RECV: Missing EOL after 5 seconds
Sep 30 10:26:29.71: [  227]: RECV: REJECT page quality, 93% good lines (95%
required)
Sep 30 10:26:29.85: [  227]: RECV: end page
Sep 30 10:26:31.88: [  227]: MODEM <Timeout>
Sep 30 10:26:32.88: [  227]: RECV FAX (00000257): recvq/fax00111.tif from
+49 9132 40599, route to <unspecified>, 0 pages in 2:01
Sep 30 10:26:33.71: [  227]: RECV FAX: T.30 T2 timeout, expected page not
received
Sep 30 10:26:33.80: [  227]: <-- [9:AT+FTH=3\r]
Sep 30 10:26:34.01: [  227]: --> [7:CONNECT]
Sep 30 10:26:35.19: [  227]: --> [2:OK]
Sep 30 10:26:35.73: [  227]: RECV FAX (00000257): session with +49 9132
40599 terminated abnormally: T.30 T2 timeout, expected page not received
Sep 30 10:26:35.82: [  227]: RECV FAX: bin/faxrcvd "recvq/fax00111.tif"
"fax" "00000257" "T.30 T2 timeout, expected page not received"

Liebe Gruesse

     Gregor
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Gregor Longariva <grlongar@cip.informatik.uni-erlangen.de>

From support@zoomtel.com  Wed Feb 17 13:55:47 1999
From: Zoom Technical Support <support@zoomtel.com>
To: john@auctor.auctorcorp.com
Subject: Re: Adaptive answer 
Date: Fri, 26 Sep 1997 06:07:48 -0400

Hi John,

It'll work but we don't support it! (Unix) I run the same model on
Linux and it works great. Try one of these Init Strings if you're
having problems...,


AT&F&C1&D2%C0S10=200&W

AT&F&C1&D2\A2%C1S10=200&W

AT&F&C1&D2\A2%C1S95=1S10=200&W

AT&F&C1&D2\N5\A2%C1&W

AT&F&C1&D2\N5&K3S95=1S10=200&W

AT&F%C1&D2&Q5S7=100S10=200\A2&W

AT&F&C1&D2\K0%C0&K3&W

AT&F&C1&D2\K0%C1S95=1&W

AT&F&C1&D2&K3MO%C1S10=200S95=44+MS=11,1,21600,33600&W

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
X-Authentication-Warning: guinness.chelsea.net: davidb owned process doing -bs
Date: Fri, 3 Oct 1997 00:01:51 -0400 (EDT)
From: David Birnbaum <davidb@chelsea.net>
Reply-To: David Birnbaum <davidb@chelsea.net>
To: flexfax@sgi.com
Subject: flexfax: Perl HylaFax Interface, and some bugs
Sender: owner-flexfax@celestial.com

Folks,

I'm in the process of finishing a Perl 5 OO interface to Hylafax.  In
another week or so, it will be stable enough to be alpha, so I can toss it
onto the net for whoever is interested in it.  I'm putting in methods for
the whole interface, piece-by-piece; still feeling out the structure, but
it seems to work for the type of stuff I want to do with it.  I'm trying
my best to use the sendfax configuration files to make it simple to use
for those who have mastered sendfax.

However, as I've been refining the interface, I've run into a few strange
bugs that appear in the HylaFax protocol, and I was hoping people would
have a few comments.  Specifically, the problems came up in the creation
of a Clone method, which would clone a job and then allow you to resubmit
it with a different dialstring, tagline, etc, for use in fax broadcasting. 

Bug #1 - LASTTIME changes very strangely: take a look at the following
snippet:

  jnew
  200 New job created: jobid: 3733 groupid: 3733.
  jparm lasttime 001440
  213 LASTTIME set to 001440.
  jparm
  217-Job state: jobid 3733 groupid 3733
      SENDTIME: NOW
      LASTTIME: 101370431

How did LASTTIME get set to this strange number?  I only noticed this
because my Clone function actually duplicates the whole job, so when it
cloned the JPARM LASTTIME 101370431, the server complained and the job
blew up.  Oddly enough, the sendq file seems to have something completely
unrelated to the number I put in as well as the number that's returned:

  killian /var/spool/fax/sendq 197# grep kill q3733
  killtime:10800

Bug #2: "Incomplete" jobs: if a job is only halfway completed, and then
the Hylafax connection is broken, you can't get back into the job to fix
it, finish it, or delete it.  If I were to QUIT out of a job after
defining only the items in the example above:

  job 3733
  500 Cannot set job 3733; invalid or corrupted job description file.

This is a major pain, because I can't recover very gracefully from a job
submission failing to finish, losing a connection, deleting a job that is
missing information, etc. This error, as well as the generic "Server has
N'ACKd the request" error, are very frustrating because they give no
indication on how to fix the problem.  Informative error messages would be
very handy here (what was corrupted?) 

Pointers, comments, etc, would be greatly appreciated here.

Bug #3: If a job fails due a modem initialization error (ie, "Can't
configure modem for faxing"), then I believe the job should be rescheduled
for another modem rather than be cancelled as it is now.  I couldn't find
any way to change this behavior in the documentation.

Thanks in advance for any assistance and comments on the above.  I've only
been on this list for a few weeks, so apologies if this stuff has already
come up and gone by.

Looking forward to pl2....

David.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 3 Oct 1997 02:13:56 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: "Steven E. Frazier" <sfrazier@futurenet.net>
Cc: "'flexfax@sgi.com'" <flexfax@sgi.com>
Subject: Re: flexfax: FW: HOWTO: Get Hylofax and Ghostscript to work together
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Thu, 2 Oct 1997, Steven E. Frazier wrote:

>   First, incoming faxes are being stored as a tiff file.  We are using
> samba to map the fax receive area to a windows share where we are trying to
> view them.  Paint shop pro says that it's not a valid tiff file.  So...

It's tiffg3. Since Hylafax was written by Sam Leffler, who invented TIFF,
he gets to call it TIFF as well.

Under Win95, there is a freeware program available called, I believe,
acdsee. Under UNIX, take a look at ftp://ftp.sgi.com/sgi/fax/contrib/
and get the viewfax package.

I've modified my "facrcvd" script under my local installation to name
the received files "*.fax" instead of "*.tif", for better web access.
You can alternately use the "fax2ps" script to turn them to Postscript,
or the "tiffcp" program in the tiff package you installed.

>   1) What viewer could I use to see/read/print the fax?  Presume that we
> don't have X running on the linux box.
>   2) What utilities exist to convert this file into a gif, jpeg or HP
> format file ? 
> (the last resort being to just print it on his HP printer).

Overall conversion can be done with the ImageMagick "convert" utility,
but I do not recommend it. Use "tiffcp" first to turn it into distinct
TIFF files, one per page, and *then* convert them to whatever.

>   Next item,  we can't send outgoing faxes because Ghostscript doesn't like
> the fonts that are being requested.   Aparently, the Hylofax package wants
> to know the outline bitmaps by names other than Linux has them.  For
> instance, in our /usr/lib/ghostscript/fonts directory, it wants to see a
> file "Courer-BOLD' and we have file names that look like serial numbers. 
> The text to fax utility dies saying that 'it can't obtain font metrics for
> 'Courier-BOLD'.  You may fill in the blank for the font name; I can't get
> it to work using any name.

You need the afm fonts. Take a look at the
ftp://ftp.sgi.com/sgi/fax/source/, and make sure you don't already
have them installed someplace odd (which can be set at configuration
time or in the config.site file for HylaFAX).

>   1)  Should I rename or add a symlink to make the names be what Hylofax
> wants to see or...
> 
>   2) Am I missing something more basic than that?

See above.

>   Third,  we can get the program to do an alpha page, but it fails to work
> with numeric only pagers.  I know my description of this problem is scanty,
> but I presume that only someone who has set up Hylofax would be able to
> answer it anyway.

Can't help you there.


			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNDSNGj/+ItycgIJRAQEzQwP/TFNkPL5VoZzOTNjNhuPmvnclbkWqOpFC
KttN54/WafF/eiVJaX+KWlmIN5b6xv3dtXXrR44aXwwQlOR2uIMW36Pi+D5C8Ind
/CJhnMJuI4yqKRJB9LFWzsJsdFP+VQ0R5i1078LsKJvI6ucDrIBnog0cyKuaGd6x
eIUYbVDXShg=
=R9q1
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 25 Sep 1997 12:45:41 -0400 (EDT)
Organization: The Creative Group
From: Patrick Gardella <patrick@cre8tivegroup.com>
To: flexfax@celestial.com
Subject: flexfax: word wrap vs truncate
Sender: owner-flexfax@celestial.com

I haven't heard from anyone on this, so I thought I would try again!  Can 
someone HELP?
____

I'm running 4.0pl1 on FreeBSD 2.2.1 (Pentium 200 128Megs RAM) with 8 Multitech
ZDK2834 Multimodems.  

HylaFAX is running just fine.  I've got it configured as an email to fax 
gateway.  However, I'm trying to get LineWrap to work right.

from man faxmail:
       LineWrap       Control whether long text lines are wrapper
                      or truncated  at  the  right  hand  margin.
                      (Equivalent to the -c option.)

       Tag               Type         Default         Description
       LineWrap          boolean      Yes            wrap/truncate text lines

Now if I set LineWrap:  Yes in /usr/local/lib/fax/hyla.conf, I get the lines
truncated.  No problem there.

If I set it to No, the lines go off the end of the page with no wrap whatsoever.
So I lose whatever didn't fit on the page.  Ok...

If I don't have LineWrap in hyla.conf at all, the lines wrap, at the right edge
of the fax, but they get cut off in the middle of the words.  It looks like at
80 characters, it cuts it off, whether or not there is a word there or not. 
Isn't it supposed to cut it at around 80 characters at word boundaries? 

My hyla.conf is:
KillTime:       "now +3 days"
MaxDials:       30
GaudyHeaders:   No
Headers:        clear
PageSize:       na-let
TextPointSize:  14pt
TextFont:       Times-Roman

Any suggestions?  Searching the mla for "wrap" "truncate" or "-c" didn't yeild
any answers.

Patrick

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 30 Sep 1997 01:37:44 -0400
From: Lothar Krau <Lothar_Krauss@compuserve.com>
Subject: flexfax: Hylafax and Creatix modem
To: Andreas Pschl <andy@bigboy.degnet.baynet.de>,
        "(unbekannt)" <flexfax@sgi.com>
Sender: owner-flexfax@celestial.com

Andreas wrote:
> I cant get my Creatix Phonemaster VF144 modem working together with Hylafax4.0 under > Linux. 
> Could you please help me to get the right initialisation codes (it has got a UMC 
> chipset).
> None of the ones I found, worked right!

Here is my config file for Creatix LC144. It works for outgoing and incomming faxes.

Lothar

CountryCode:            49
AreaCode:               69
FAXNumber:              +xx.xx.xxxxxxx
LongDistancePrefix:     0
InternationalPrefix:    00
DialStringRules:        etc/dialrules
ServerTracing:          1
SessionTracing:         63
RecvFileMode:           0600
LogFileMode:            0600
DeviceMode:             0600
RingsBeforeAnswer:      1
SpeakerVolume:          Medium
GettyArgs:              "%l"
LocalIdentifier:        "xxxxxxxxxxxxxxx"
TagLineFont:            etc/lutRS18.pcf
TagLineFormat:          "Von %%l|%%n|%d.%m.%Y %R|Seite %%p von %%t"
QualifyTSI:             ""
MaxRecvPages:           25
ModemType:              Class2          # use this to supply a hint
ModemRate:              19200           # max rate for DCE-DTE communication
ModemFlowControl:       xonxoff         # XON/XOFF flow control assumed
ModemSetupDTRCmd:       AT&D2           # setup so DTR drop resets modem
ModemSetupDCDCmd:       AT&C1           # setup so DCD reflects carrier (or not)
ModemDialCmd:           AT%%TCBDT%s     # Keine Wahlsperre bei Fehlversuchen
ModemResetCmds:         AT+FAA=1        # stuff to do when modem is reset
ModemResetDelay:        3500
ModemAnswerCmd:         ATA             # use this to answer phone
ModemNoFlowCmd:         AT&K0           # disable flow control cmd
ModemHardFlowCmd:       AT&K3           # hardware flow control cmd
ModemSoftFlowCmd:       AT&K4           # software flow control cmd
ModemNoAutoAnswerCmd:   ATS0=0          # disable auto-answer
ModemWaitForConnect:    true            # User-Name CONNECT existiert nicht
ModemSetVolumeCmd:      "ATM0 ATL0M1 ATL1M1 ATL2M1 ATL3M1"
ModemEchoOffCmd:        ATE0            # disable command echo
ModemVerboseResultsCmd: ATV1            # enable verbose command results
ModemResultCodesCmd:    ATQ0            # enable result codes
ModemOnHookCmd:         ATH0            # place phone on hook (hangup)
ModemSoftResetCmd:      ATZ             # do soft reset of modem
ModemWaitTimeCmd:       ATS7=60         # wait 60 seconds for carrier
ModemCommaPauseTimeCmd: ATS8=2          # comma pause time is 2 seconds
ModemRecvFillOrder:     MSB2LSB         # bit order of received facsimile
ModemSendFillOrder:     LSB2MSB         # bit order modem expects for transmit
Use2D:                  No
Class2Cmd:              AT+FCLASS=2     # command to enter class 2
Class2BORCmd:           AT+FBOR=0       # bit order for phase B/C/D (direct)
Class2RELCmd:           AT+FREL=1       # byte-align EOL codes on recv
Class2CQCmd:            ""              # commands to enable copy quality proc.
Class2AbortCmd:         AT+FK           # abort session command
Class2CQQueryCmd:       AT+FCQ=?        # query for copy quality capabilities
Class2DCCQueryCmd:      AT+FDCC=?       # query for modem capabilities
Class2TBCCmd:           AT+FTBC=0       # setup stream mode
Class2CRCmd:            AT+FCR=1        # enable capability to receive
Class2PHCTOCmd:         AT+FPHCTO=30    # set phase C timeout parameter
Class2BUGCmd:           AT+FBUG=1       # enable HDLC frame tracing
Class2LIDCmd:           AT+FLID         # set local identifier command
Class2DCCCmd:           AT+FDCC         # set modem capabilities command
Class2DISCmd:           AT+FDIS         # set session parameters command
Class2DDISCmd:          AT+FDIS         # Disable pre-dial DIS command hack
Class2CIGCmd:           AT+FCIG         # set polling identifier command
Class2PTSCmd:           AT+FPTS         # set post-page status command
Class2SPLCmd:           AT+FSPL         # set polling indicator command
Class2RecvDataTrigger:  "\022"          # character sent to modem to start recv
Class2XmitWaitForXON:   no              # if true, wait for XON before send
From: jay@dixonssurgical.co.uk (Jay Dixon)
To: flexfax@sgi.com
Subject: flexfax: Setup Problems
Date: Wed, 01 Oct 1997 16:11:25 GMT
Organization: Dixons Surgical Ltd
Reply-To: jay@dixonssurgical.co.uk
X-Mailer: Forte Agent 1.5/32.451
Sender: owner-flexfax@celestial.com

I am trying to set up a Hylafax v4.0pl1 server on FreeBSD 2.2.1 to
serve outgoing only for Win3.x/WinNT4.0 machines.  I have a Hayes
ACCURA 56K External Speakerphone Modem on cuaa0.  This is a class 1
fax device.  The cuaa0 dev is also used for PPP dialup to my ISP using
userPPP within FreeBSD.

I have got the software installed but the setup is still shaky.  I
have the hfaxd service setup in inetd on port 4559 and this is working
OK from boot (i.e. the Win machines can communicate and pass faxes
through).

I am starting faxq from the following script at bootup in
/usr/local/etc/rc.d/hylafax.sh -

if [ -x /usr/local/sbin/faxq ]; then
        /usr/local/sbin/faxq && echo 'Hylafax Server.'
fi

The faxq starts up OK but when I send a fax through it sits there
saying 'Modem cuaa0 (+44.(0)1268.764615): Waiting for modem to come
ready'.

If I then type '/usr/local/sbin/faxmodem cuaa0' at the command line
the modem then dials and sends the faxes OK.  What am I missing and
where?

Also after the modem has sent the fax it is staying in class1 mode and
I can't send data (i.e. make a PPP connection).  I have tried the
following strings in the 'ModemResetCmds' variable of
/var/spool/fax/etc/config.cuaa0 with no joy -

AT+FCALSS=0	# Set to Fax Class 0 (Data)
ATF		# Set to factory presets

Any ideas please?

Thanks
Jay

---------
Jay Dixon - Dixons Surgical Ltd - Wickford - Essex - England
jay@dixonssurgical.co.uk  -  http://www.dixonssurgical.co.uk

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 1 Oct 1997 16:39:30 -0400 (EDT)
From: Brendon Perero <perero@wopr.inetu.net>
To: Jay Dixon <jay@dixonssurgical.co.uk>
cc: flexfax@sgi.com
Subject: Re: flexfax: Setup Problems
Sender: owner-flexfax@celestial.com

Try starting the faxq with /usr/local/sbin/faxq -m cuaa0

Brendon


On Wed, 1 Oct 1997, Jay Dixon wrote:

> I am trying to set up a Hylafax v4.0pl1 server on FreeBSD 2.2.1 to
> serve outgoing only for Win3.x/WinNT4.0 machines.  I have a Hayes
> ACCURA 56K External Speakerphone Modem on cuaa0.  This is a class 1
> fax device.  The cuaa0 dev is also used for PPP dialup to my ISP using
> userPPP within FreeBSD.
> 
> I have got the software installed but the setup is still shaky.  I
> have the hfaxd service setup in inetd on port 4559 and this is working
> OK from boot (i.e. the Win machines can communicate and pass faxes
> through).
> 
> I am starting faxq from the following script at bootup in
> /usr/local/etc/rc.d/hylafax.sh -
> 
> if [ -x /usr/local/sbin/faxq ]; then
>         /usr/local/sbin/faxq && echo 'Hylafax Server.'
> fi
> 
> The faxq starts up OK but when I send a fax through it sits there
> saying 'Modem cuaa0 (+44.(0)1268.764615): Waiting for modem to come
> ready'.
> 
> If I then type '/usr/local/sbin/faxmodem cuaa0' at the command line
> the modem then dials and sends the faxes OK.  What am I missing and
> where?
> 
> Also after the modem has sent the fax it is staying in class1 mode and
> I can't send data (i.e. make a PPP connection).  I have tried the
> following strings in the 'ModemResetCmds' variable of
> /var/spool/fax/etc/config.cuaa0 with no joy -
> 
> AT+FCALSS=0	# Set to Fax Class 0 (Data)
> ATF		# Set to factory presets
> 
> Any ideas please?
> 
> Thanks
> Jay
> 
> ---------
> Jay Dixon - Dixons Surgical Ltd - Wickford - Essex - England
> jay@dixonssurgical.co.uk  -  http://www.dixonssurgical.co.uk
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 2 Oct 1997 12:13:32 +0200 (CEST)
From: Thomas Pickel <tp@gandalf.franken.de>
To: flexfax@sgi.com
Subject: flexfax: Error after successfully receiving fax documents
Sender: owner-flexfax@celestial.com


I'm using hylafax 4.0pl1 on Linux 2.0.29 with a Creatix Modem
(ROCKWELL AC/V34/V1.510-V34_DS). Sending works fine, and I can also
receive faxes. The problem is that all fax machines that send me faxes
think there was an error ("Unable to detect handshake signal for next
page") though the page was received perfectly. They then retransmit the
fax over and over.

Has anybody experienced the same problem and found a solution?

Thomas Pickel
tp@gandalf.franken.de

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 3 Oct 1997 16:24:45 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Guascor I+D <guascor_id@jet.es>
Cc: Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: How to force pulse dialing (ATDP)
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Fri, 3 Oct 1997, Guascor I+D wrote:

> I have installed hylafax latest version on two Sun sparc 5 running solaris
> 2.5.1. I got 2 Us Robotics Sporters voice 33.6 Faxmodem and Ive read the
> HylaFax FAQ and trobleshooting many times. These machines are located in
> two diferent locations, the only diference between them is that one uses
> tone dialing while the other uses pulse dialing. My problem is that i don't
> know how to tell hylafax to dial usin ATDP instead of ATDT. )
> I've changed every single file containing the text ATDT (well, actually i
> only found those files under etc and config ), but when i send a fax it
> says NO CARRIER DETECTED and looking at the log file i see it stills dials
> using ATDT.

Hmm. Re-read the manual page for the "config" files, set

	ModemDialCmd	ATDT%s@		# T for Tone dialing, @ for silence

to the appropriate value for your modem. Then run "faxsetup" to restart
your daemons so they read the new config files.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNDVUgD/+ItycgIJRAQFokgQAkcDd41uDlYOx/nvcQudcfyKIL0KsqAP8
nEuVqjas5e38lFf0sgQRWsviu3jAS4uPx0YmkMTQZQcrHWih+7HZUJtpi6xbW15O
Pr+97nkX++mo+++kO37rlrawODXd+ol6E5v3bQ0fTniYA7pMdhGECfL93kDZXiGT
YOc9Qa1sTng=
=BklV
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: SendMail Problem
To: juliochegedus@planethouse.com.br
Date: Fri, 3 Oct 1997 23:44:58 +0100 (BST)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> Oct 1 12:12:12 host sendmail [78]: PAA00353: SYSERR(root): mailer prog
> died with signal 13
> -----------
> 
> In sendmail's documentation I don't find nothing related, and sendmail

You wouldn't expect to; the error is self explanatory as far as sendmail 
is concerned, and it has told you all it knows.

At the very least, we need to know your operating system, but even better
would be to know what signal 13 is.  (On Linux, it is SIGPIPE, which means
that a write to a pipe failed because a downstream reader closed it
prematurely.  It is also SIGPIPE on SCO Unix, so it is probably fairly
standard.)

Generally sendmail will include error output from the failing program
in the bounce message, if delivery is deferred, or on the terminal
if delivery is immediate.  That is probably where the real diagnostic
will be.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: Wrong destination
To: ameissonnier@dohle.com
Date: Sat, 4 Oct 1997 10:32:16 +0100 (BST)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> Oct 01 16:17:14.92: [26705]: <-- [22:ATX3DT0,001512xxxxxxx\r]
                                         ^^

It looks like you are dialling the number blind.  If the PABX is 
particularly overloaded, it might not be ready to accept the number 
in time.  If you really need the comma, your PABX must be very old.

In any case, it looks like a modem, or modem configuration problem.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 06 Oct 1997 18:47:50 +0100
To: HylaFAX List <flexfax@sgi.com>
From: Phil Watkinson <pkw@elgro.co.uk>
Subject: flexfax: Any additions to the TODO List?
Sender: owner-flexfax@celestial.com


Hi.

I am currently undertaking a minor revision to the html documentation as
part of the pl2 release. One task has been to add the TODO List to the
online docs, which, since there's quite a few entries, I have limited the
page to those marked as 'high priority'.

This email is to ask if anyone has new items to put to the list ? (or any
amendments/deletions generally)?

Could I start the ball rolling with the following 2 wishes :-

1. Textfmt to be able to recognize and use ghostscript's Fontmap.

  [At the moment, users have to use the 'afm fonts' instead or create 
   links directly to the ghostscript font files, although there is a 
   perl script that at least does this now ].

2. The capability to configure HylaFAX via a web browser. And perhaps
   extended to cover fax status and statistics as well.

  [This was raised in the recent 'is HylaFAX's future a NT port?' thread
   where it was proposed by Davide Bolcioni as a way to build on Unixs'
   traditional strengths with the perceived strengths of MS NT].

I am realistic about the chances of this actually happening, but still
think it helpful to mention it in the TODO List (if agreed) as a 
direction to aim for. (The Marketing Department would perhaps hype it as
a 'roadmap').

Any comments?

Cheers,
Phil Watkinson,
Boston, UK.
    

From neil@vic.design.telstra.com.au  Wed Feb 17 13:55:47 1999
To: Phil Watkinson <pkw@elgro.co.uk>, flexfax@sgi.com
Subject: Re: flexfax: Any additions to the TODO List? 
Date: Tue, 07 Oct 1997 11:33:16 +1000
From: Neil Buckingham <neil@vic.design.telstra.com.au>

	Here is another item for the todo list.

	hylafax v4.0pl1 IXO/TAP pager protocol not conforming to Australian ( 
Probably international also ) standard.
	When sending a pager message to a invalid pager number, hylafax just drops 
the line and does not go through the correct disconnect procedures. It does do 
the correct disconnect procedures when sending to a valid pager number.

	e.g.
	Here is a trace from a valid pager number

Oct 07 09:20:49.34: [23526]: EXPECT ID (paging central identification)
Oct 07 09:20:51.35: [23526]: MODEM TIMEOUT: reading line from modem
Oct 07 09:20:51.70: [23526]: RECV ID ("ID=")
Oct 07 09:20:51.70: [23526]: SEND device identification/login request
Oct 07 09:20:52.44: [23526]: RECV:  Log-on accepted
Oct 07 09:20:52.44: [23526]: RECV ACK (login successful)
Oct 07 09:20:52.45: [23526]: RECV: Ready for messages
Oct 07 09:20:52.45: [23526]: RECV go-ahead (prologue done)
Oct 07 09:20:52.45: [23526]: SEND message block
Oct 07 09:20:53.13: [23526]: RECV: Send next message
Oct 07 09:20:53.13: [23526]: RECV ACK (message block accepted)
Oct 07 09:20:53.65: [23526]: RECV: Thank you for calling
Oct 07 09:20:53.66: [23526]: RECV EOT (disconnect)
Oct 07 09:20:53.66: [23526]: SEND PAGE: FROM www@zarquon.vic.design.telecom.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Any additions to the TODO List?
To: neil@vic.design.telstra.com.au (Neil Buckingham)
Date: Tue, 7 Oct 1997 08:41:32 +0200 (MSZ)
Cc: flexfax@sgi.com, pkw@elgro.co.uk (Phil Watkinson)
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Neil Buckingham wrote:
    
    	Here is another item for the todo list.
    
    	hylafax v4.0pl1 IXO/TAP pager protocol not conforming to Australian ( 
    Probably international also ) standard.
    	When sending a pager message to a invalid pager number, hylafax just drops 
    the line and does not go through the correct disconnect procedures. It does do 
    the correct disconnect procedures when sending to a valid pager number.
    
    	e.g.
    	Here is a trace from a valid pager number

	...

Please provide a log for a invalid pager number with the trace
of low level IXO/TAP protocol enabled. Thanks in advance.

	matthias

From neil@vic.design.telstra.com.au  Wed Feb 17 13:55:47 1999
To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
Cc: neil@orca1.vic.design.telstra.com.au (Neil Buckingham), flexfax@sgi.com,
        pkw@elgro.co.uk (Phil Watkinson)
Subject: Re: flexfax: Any additions to the TODO List? 
Date: Fri, 10 Oct 1997 08:09:52 +1000
From: Neil Buckingham <neil@vic.design.telstra.com.au>

> Neil Buckingham wrote:
>     
>     	Here is another item for the todo list.
>     
>     	hylafax v4.0pl1 IXO/TAP pager protocol not conforming to Australian ( 
>     Probably international also ) standard.
>     	When sending a pager message to a invalid pager number, hylafax just drops 
>     the line and does not go through the correct disconnect procedures. It does do 
>     the correct disconnect procedures when sending to a valid pager number.
>     
>     	e.g.
>     	Here is a trace from a valid pager number
> 
> 	...
> 
> Please provide a log for a invalid pager number with the trace
> of low level IXO/TAP protocol enabled. Thanks in advance.
> 
> 	matthias


	Sorry, I had a trace from a invalid pager but it seems to have been chopped 
off. Anyway here is another one.


Oct 10 07:36:25.95: [19240]: SESSION BEGIN 00003295 123456789
Oct 10 07:36:25.95: [19240]: SEND PAGE: JOB 2663 DEST 123456789 COMMID 00003295
Oct 10 07:36:25.95: [19240]: MODEM set DTR OFF
Oct 10 07:36:25.95: [19240]: DELAY 2600 ms
Oct 10 07:36:28.55: [19240]: MODEM set DTR ON
Oct 10 07:36:28.55: [19240]: MODEM set baud rate: 19200 baud, input flow 
XON/XOF
F, output flow XON/XOFF
Oct 10 07:36:28.55: [19240]: MODEM flush i/o
Oct 10 07:36:28.55: [19240]: <-- [15:ATE0V1Q0S0=0H0\r]
Oct 10 07:36:29.72: [19240]: --> [2:OK]
Oct 10 07:36:29.72: [19240]: <-- [21:ATS8=2S7=60&K4&D3&C1\r]
Oct 10 07:36:29.84: [19240]: --> [2:OK]
Oct 10 07:36:29.84: [19240]: <-- [5:ATM0\r]
Oct 10 07:36:29.94: [19240]: --> [2:OK]
Oct 10 07:36:29.94: [19240]: MODEM set parity: 7 bits, even parity
Oct 10 07:36:29.94: [19240]: MODEM input buffering enabled
Oct 10 07:36:29.96: [19240]: <-- [12:AT+FCLASS=0\r]
Oct 10 07:36:30.16: [19240]: --> [2:OK]
Oct 10 07:36:30.16: [19240]: DIAL 95104451
Oct 10 07:36:30.16: [19240]: <-- [13:ATDT1234567\r]
Oct 10 07:36:40.71: [19240]: --> [7:RINGING]
Oct 10 07:36:54.14: [19240]: --> [12:CONNECT 2400]
Oct 10 07:36:54.14: [19240]: EXPECT ID (paging central identification)
Oct 10 07:36:54.14: [19240]: <-- <1:\015>
Oct 10 07:36:56.15: [19240]: MODEM TIMEOUT: reading line from modem
Oct 10 07:36:56.15: [19240]: <-- <1:\015>
Oct 10 07:36:56.49: [19240]: --> <3:ID=>
Oct 10 07:36:56.49: [19240]: RECV ID ("ID=")
Oct 10 07:36:56.49: [19240]: SEND device identification/login request
Oct 10 07:36:56.49: [19240]: <-- <11:\033PG1pwpw\015>
Oct 10 07:36:57.23: [19240]: --> <39: Log-on accepted                       >
Oct 10 07:36:57.23: [19240]: RECV:  Log-on accepted
Oct 10 07:36:57.23: [19240]: --> <1:\006>
Oct 10 07:36:57.23: [19240]: RECV ACK (login successful)
Oct 10 07:36:57.23: [19240]: --> <39:Ready for messages                     >
Oct 10 07:36:57.23: [19240]: RECV: Ready for messages
Oct 10 07:36:57.23: [19240]: --> <3:\033[p>
Oct 10 07:36:57.23: [19240]: RECV go-ahead (prologue done)
Oct 10 07:36:57.23: [19240]: SEND message block
Oct 10 07:36:57.23: [19240]: <-- <22:\0021234\015hello test\015\0034==\015>
Oct 10 07:36:57.83: [19240]: --> <39:Pager Id 1234    is not valid          >
Oct 10 07:36:57.83: [19240]: RECV: Pager Id 1234    is not valid
Oct 10 07:36:57.83: [19240]: --> <1:\036>
Oct 10 07:36:57.83: [19240]: RECV RS (message block rejected; skip to next)
Oct 10 07:36:57.83: [19240]: <-- [5:ATH0\r]
Oct 10 07:37:02.84: [19240]: MODEM TIMEOUT: reading line from modem
Oct 10 07:37:02.84: [19240]: MODEM <Timeout>
Oct 10 07:37:02.84: [19240]: MODEM input buffering enabled
Oct 10 07:37:02.86: [19240]: MODEM set DTR OFF
Oct 10 07:37:02.86: [19240]: SESSION END


And here is a valid address

Oct 10 07:41:28.02: [19282]: SESSION BEGIN 00003296 123456789
Oct 10 07:41:28.02: [19282]: SEND PAGE: JOB 2664 DEST 123456789 COMMID 00003296
Oct 10 07:41:28.02: [19282]: MODEM set DTR OFF
Oct 10 07:41:28.02: [19282]: DELAY 2600 ms
Oct 10 07:41:30.61: [19282]: MODEM set DTR ON
Oct 10 07:41:30.61: [19282]: MODEM set baud rate: 19200 baud, input flow 
XON/XOF
F, output flow XON/XOFF
Oct 10 07:41:30.61: [19282]: MODEM flush i/o
Oct 10 07:41:30.61: [19282]: <-- [15:ATE0V1Q0S0=0H0\r]
Oct 10 07:41:31.78: [19282]: --> [2:OK]
Oct 10 07:41:31.78: [19282]: <-- [21:ATS8=2S7=60&K4&D3&C1\r]
Oct 10 07:41:31.90: [19282]: --> [2:OK]
Oct 10 07:41:31.90: [19282]: <-- [5:ATM0\r]
Oct 10 07:41:32.01: [19282]: --> [2:OK]
Oct 10 07:41:32.01: [19282]: MODEM set parity: 7 bits, even parity
Oct 10 07:41:32.01: [19282]: MODEM input buffering enabled
Oct 10 07:41:32.02: [19282]: <-- [12:AT+FCLASS=0\r]
Oct 10 07:41:32.22: [19282]: --> [2:OK]
Oct 10 07:41:32.22: [19282]: DIAL 95104451
Oct 10 07:41:32.22: [19282]: <-- [13:ATDT1234567\r]
Oct 10 07:41:42.79: [19282]: --> [7:RINGING]
Oct 10 07:41:56.21: [19282]: --> [12:CONNECT 2400]
Oct 10 07:41:56.21: [19282]: EXPECT ID (paging central identification)
Oct 10 07:41:56.21: [19282]: <-- <1:\015>
Oct 10 07:41:58.22: [19282]: MODEM TIMEOUT: reading line from modem
Oct 10 07:41:58.22: [19282]: <-- <1:\015>
Oct 10 07:41:58.57: [19282]: --> <3:ID=>
Oct 10 07:41:58.57: [19282]: RECV ID ("ID=")
Oct 10 07:41:58.57: [19282]: SEND device identification/login request
Oct 10 07:41:58.57: [19282]: <-- <11:\033PG1pwpw\015>
Oct 10 07:41:59.32: [19282]: --> <39: Log-on accepted                       >
Oct 10 07:41:59.32: [19282]: RECV:  Log-on accepted
Oct 10 07:41:59.32: [19282]: --> <1:\006>
Oct 10 07:41:59.32: [19282]: RECV ACK (login successful)
Oct 10 07:41:59.32: [19282]: --> <39:Ready for messages                     >
Oct 10 07:41:59.32: [19282]: RECV: Ready for messages
Oct 10 07:41:59.32: [19282]: --> <3:\033[p>
Oct 10 07:41:59.32: [19282]: RECV go-ahead (prologue done)
Oct 10 07:41:59.32: [19282]: SEND message block
Oct 10 07:41:59.32: [19282]: <-- <25:\0026388040\015hello test\015\003580\015>
Oct 10 07:41:59.96: [19282]: --> <39:Send next message                      >
Oct 10 07:41:59.96: [19282]: RECV: Send next message
Oct 10 07:41:59.96: [19282]: --> <1:\006>
Oct 10 07:41:59.96: [19282]: RECV ACK (message block accepted)
Oct 10 07:41:59.96: [19282]: <-- <2:\004\015>
Oct 10 07:42:00.48: [19282]: --> <39:Thank you for calling                  >
Oct 10 07:42:00.48: [19282]: RECV: Thank you for calling
Oct 10 07:42:00.48: [19282]: --> <2:\033\004>
Oct 10 07:42:00.48: [19282]: RECV EOT (disconnect)
Oct 10 07:42:00.48: [19282]: SEND PAGE: FROM root@brainiac.vic.design.telecom.c
o
m.au TO 95104451 (sent in 0:28)
Oct 10 07:42:00.48: [19282]: <-- [5:ATH0\r]




This problem causes it to fail the test that Telstra does ony any pager 
software connect to their system, and so it cannot be officially used.



While I am at it one more wish.
	Not sure how to explain this. But hylafax queues jobs for the same phone 
number and will only send one at a time. This make sense for a fax job. 
However pager jobs are all sent to the same phone number but otherwise are 
unreleated. hylafax should be able to send pager jobs in parallel. i.e. if you 
have 10 phone lines and 10 pager jobs in the queue then send the jobs in 
parallel using all the modems. As you can probably tell we are mainly using 
hylafax as a paging service. And so the way it is at the moment there is no 
point in having more than one modem line as it won't send pager jobs in 
parallel.



Thanks

Neil Buckingham.




-- 
----------------------------- /-_|\ -----------------------------------
Neil Buckingham              /     \         5th Floor 360 Elizabeth St
IBM GSA                      \_,-._/         Melbourne Australia 3000
Network Systems                   v          Phone +613 9203 7455
EMail n.buckingham@vic.design.telecom.com.au Fax +613 9663 8229
PGP Key fingerprint =  74 F2 64 A6 1C E0 48 8D  85 3C A4 A0 86 0C 85 0E 
-----------------------------------------------------------------------
"A coward is a frightened man who's scared to be brave. But a brave
man is only a coward who isn't scared to be frightened."
-- Maxwell Smart. Get Smart.

From Matthias.Apitz@SOFTCON.de  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Any additions to the TODO List?
To: neil@vic.design.telstra.com.au (Neil Buckingham)
Date: Sun, 19 Oct 1997 18:14:09 +0200 (MSZ)
Cc: pkw@elgro.co.uk (Phil Watkinson)
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps

Neil Buckingham wrote:
    
    > 	...
    > 
    > Please provide a log for a invalid pager number with the trace
    > of low level IXO/TAP protocol enabled. Thanks in advance.
    > 
    > 	matthias
    
    
    	Sorry, I had a trace from a invalid pager but it seems to have been chopped 
    off. Anyway here is another one.
    
    
    <....trace deleted ...>

Could you please test a small change? In faxd/pageSendApp.c++ line
300 ... says:

            if (req.status == send_ok)
                (void) pageEpilogue(req, info, notice);
        } else
            .....

Please remove the "if (req.status == send_ok)" or comment it out,
run a test and examine the logs carefully. The pageEpilogue()
will do the end-of-transmission msg (<EOT><CR>) but will also
check the response from the provider which should be <ESC><EOT>.
In case of missing <ESC><EOT> from the provider it will set
req.status=send_retry which will retry the job later.

I can't do tests here in Germany with our GSM provider D1-TELEKOM.
D1 is allways returning ACK (even if the PIN is wrong).

	matthias
-- 
firm: matthias.apitz@sisis.de [voc:+49-89-61308-351, fax: +49-89-61308-188]
priv: guru@thias.muc.de
 PGP: Key fingerprint = 0C 01 F2 23 EC 17 A2 D5  46 2D 29 4C 0E 8B 7E 8F

from USENET:
People who run servers understand that flashy interactive interfaces have
nothing to do with the underlying functionality and often get in the way.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 6 Oct 1997 16:33:58 +0100 (MET)
From: Andreas Busse <ab@ivm.net>
To: flexfax@sgi.com
cc: Andreas Busse <ab@ivm.net>
Subject: flexfax: Problems with MPS (more pages, same document)
Sender: owner-flexfax@celestial.com


Hello there,

I'm using the latest HylaFax (hylafax-v4.0pl1) on a Linux 2.0.27 box
together with a Zyxel 1496EG+ and have problems with sending
multi-page documents to some other fax machines. Here's a part
of the debug output:

Oct 06 15:13:06.12: [12876]: SEND begin page
Oct 06 15:13:44.59: [12876]: SENT 46502 bytes of data
Oct 06 15:13:44.59: [12876]: SEND 1D RTC
Oct 06 15:13:44.59: [12876]: SEND end page
Oct 06 15:13:44.59: [12876]: SEND send MPS (more pages, same document)
Oct 06 15:14:00.28: [12876]: --> [5:ERROR]
Oct 06 15:14:00.28: [12876]: SEND recv RTN (retrain negative)
Oct 06 15:14:00.28: [12876]: <-- [7:AT+FDT\r]
Oct 06 15:14:06.11: [12876]: --> [20:+FCS:1,3,0,2,0,0,0,3]
Oct 06 15:14:06.11: [12876]: --> [7:CONNECT]
Oct 06 15:14:06.11: [12876]: SEND begin page
Oct 06 15:14:44.58: [12876]: SENT 46502 bytes of data
Oct 06 15:14:44.58: [12876]: SEND 1D RTC
Oct 06 15:14:44.58: [12876]: SEND end page
Oct 06 15:14:44.58: [12876]: SEND send MPS (more pages, same document)
Oct 06 15:15:00.25: [12876]: --> [5:ERROR]
Oct 06 15:15:00.25: [12876]: SEND recv RTN (retrain negative)

I.e, Hylafax sends the first page, send MPS, gets an error message
back and retransmits the same page, resulting in the same error
again. And so on. As you can imagine, users with such fax machines
complain about getting the same page 9 times (3 times for each trial).
Even worse, the first page is a cover page only, so these users
never see the real message text.

If you have any suggestions, please let me know. I'm not subscribed
to list, so please mail me directly!

With many thanks in advance,
Andy


--
-------------------------------------------------------------- ab@ivm.net ---
  | \    / |\  /|  Gesellschaft fuer   | Andreas Busse
  |  \  /  | \/ |  Internet,           | Im Hufen Boden 16, D-53498 Waldorf
  |   \/   |    |  Vernetzung und      | tel 02636-97690 fax 02636-9769-999
                   Mehrwertdienste mbH | info@ivm.net - http://www.ivm.net/
-----------------------------------------------------------------------------

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
To: Andreas Busse <ab@ivm.net>
Cc: flexfax@sgi.com, robn@verdi.et.tudelft.nl
Subject: Re: flexfax: Problems with MPS (more pages, same document) 
Date: Mon, 06 Oct 1997 21:23:00 +0200
From: Rob van Nieuwkerk <robn@verdi.et.tudelft.nl>
Sender: owner-flexfax@celestial.com


Hi Andreas, 

> I'm using the latest HylaFax (hylafax-v4.0pl1) on a Linux 2.0.27 box
> together with a Zyxel 1496EG+ and have problems with sending
> multi-page documents to some other fax machines. Here's a part

I have a similar problem with my Zyxel 1496E.  It is impossible to send
more than 1 page to some fax machines.  It seems there is no response
to the "MPS" command from the other side.  This happens *always* with
some fax machines.  With other macines there is *never* a problem.

I don't have any problem with *incoming* faxes.

My Setup: Linux 2.0.30, Red Hat 4.2, Zyxel 1496E (rom v6.15),
	  HylaFAX v4.0pl1 with "zyxel-1496e-2.0" config file

I think I did not have this problem with an older version of Hyla/flex fax
(v3.* ?).

	Greetings,
	Rob van Nieuwkerk

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: "Bruce M. Hahne" <hahne@aisol.net>
Subject: Re: AW: flexfax: Problems with MPS (more pages, same document)
To: flexfax@sgi.com
Date: Mon, 6 Oct 1997 15:36:11 +0900 (JST)
Cc: thomas.tornblom@hax.se, hahne@goemon.aisol.net (Bruce M. Hahne),
        support@zyxel.com
X-MIME-Autoconverted: from 8bit to quoted-printable by goemon.aisol.net id PAA00763
Sender: owner-flexfax@celestial.com

Thomas Tornblom <Thomas.Tornblom@Hax.SE> writes:

> Add:
> 
> Class2DCCQueryCmd:     "!(0,1),(0-5),(0-2),(0-2),(0,1),(0),(0),(0-7)"
> 
> to the config file. This disables the ECM support for the modem. I'm not
> sure if ZyXEL ever has gotten around to fix this. I believe the bug is still
> present, both for the 1496 and 2864 series modem.

A config line like the above is already present in the zyxel-1496e-2.0 config
file which comes with Hylafax.  So ECM is disabled by default.  I don't
believe that it's an ECM problem.  We used the default Zyxel config file
(ECM disabled) and still saw the problem noted by the author below: first
page sent multiple times, followed by an abort.  We have seen this behavior
for several destination fax numbers.  It tends, as one might imagine, to get
the recipient very upset when they get 9 or so copies of the same first page,
and nothing else.

I gave Zyxel support one of the several destination fax numbers which was
giving us problems and their response was "problem noted; switch to class 1
for that destination number".  That's not acceptable; this is an automated
system and we can't just flip back and forth between class 2.0 and class 1
on a whim.

Our solution was to change modem vendors.  For outbound faxes we now use
modems with a Lucent chipset in them.  (But I'd switch back if I could get a
working fix...)

Bruce Hahne                                  TEL: 1-408-248-2275
Access Internet Solutions Inc.               FAX: 1-408-248-2290
Worldwide Internet fax services              Web: www.aisol.net
Santa Clara, CA                              Email: hahne@aisol.net




> Care to comment ZyXEL support?
> 
> Thomas
> 
> > 
> > Hi there,
> > 
> > this seems to be a problem of the ZyXEL 1496 modem series. I have a 1496E which 
> behaves exactly the
> > same when communicating with some fax machines. Does anybody know how to fix 
> this without changing 
> > the modem? Has anybody tried using a newer firmware revision? 
> > 
> > 	Alexandre
> > --
> > Alexandre Meissonnier
> > Dohle Handelsgruppe Systemberatung GmbH
> > 53719 Siegburg (Germany)
> > ameissonnier@dohle.com
> > 
> > -----Ursprngliche Nachricht-----
> > Von:	Andreas Busse [SMTP:ab@ivm.net]
> > Gesendet am:	Montag, 6. Oktober 1997 17:34
> > An:	flexfax@sgi.com
> > Cc:	Andreas Busse
> > Betreff:	flexfax: Problems with MPS (more pages, same document)
> > 
> > 
> > Hello there,
> > 
> > I'm using the latest HylaFax (hylafax-v4.0pl1) on a Linux 2.0.27 box
> > together with a Zyxel 1496EG+ and have problems with sending
> > multi-page documents to some other fax machines. Here's a part
> > of the debug output:
> > 
> > Oct 06 15:13:06.12: [12876]: SEND begin page
> > Oct 06 15:13:44.59: [12876]: SENT 46502 bytes of data
> > Oct 06 15:13:44.59: [12876]: SEND 1D RTC
> > Oct 06 15:13:44.59: [12876]: SEND end page
> > Oct 06 15:13:44.59: [12876]: SEND send MPS (more pages, same document)
> > Oct 06 15:14:00.28: [12876]: --> [5:ERROR]
> > Oct 06 15:14:00.28: [12876]: SEND recv RTN (retrain negative)
> > Oct 06 15:14:00.28: [12876]: <-- [7:AT+FDT\r]
> > Oct 06 15:14:06.11: [12876]: --> [20:+FCS:1,3,0,2,0,0,0,3]
> > Oct 06 15:14:06.11: [12876]: --> [7:CONNECT]
> > Oct 06 15:14:06.11: [12876]: SEND begin page
> > Oct 06 15:14:44.58: [12876]: SENT 46502 bytes of data
> > Oct 06 15:14:44.58: [12876]: SEND 1D RTC
> > Oct 06 15:14:44.58: [12876]: SEND end page
> > Oct 06 15:14:44.58: [12876]: SEND send MPS (more pages, same document)
> > Oct 06 15:15:00.25: [12876]: --> [5:ERROR]
> > Oct 06 15:15:00.25: [12876]: SEND recv RTN (retrain negative)
> > 
> > I.e, Hylafax sends the first page, send MPS, gets an error message
> > back and retransmits the same page, resulting in the same error
> > again. And so on. As you can imagine, users with such fax machines
> > complain about getting the same page 9 times (3 times for each trial).
> > Even worse, the first page is a cover page only, so these users
> > never see the real message text.
> > 
> > If you have any suggestions, please let me know. I'm not subscribed
> > to list, so please mail me directly!
> > 
> > With many thanks in advance,
> > Andy
> > 
> > 
> > --
> > -------------------------------------------------------------- ab@ivm.net ---
> >   | \    / |\  /|  Gesellschaft fuer   | Andreas Busse
> >   |  \  /  | \/ |  Internet,           | Im Hufen Boden 16, D-53498 Waldorf
> >   |   \/   |    |  Vernetzung und      | tel 02636-97690 fax 02636-9769-999
> >                    Mehrwertdienste mbH | info@ivm.net - http://www.ivm.net/
> > -----------------------------------------------------------------------------
> > 
> > 
> > 
> 
> Real life:      Thomas Trnblom             Email:  Thomas.Tornblom@Hax.SE
> Snail mail:     HB Hax                      Phone:    +46  18 290 290
>                 Banvallsvgen 14            Fax:      +46  18 290 291
>                 S - 754 40 Uppsala, Sweden  Cellular: +46 708 323118

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: "Alexandre Meissonnier (home)" <ameissonnier@dohle.com>
To: "'Andreas Busse'" <ab@ivm.net>
Cc: "'flexfax@sgi.com'" <flexfax@sgi.com>
Subject: AW: flexfax: Problems with MPS (more pages, same document)
Date: Mon, 6 Oct 1997 20:49:48 +0200
Sender: owner-flexfax@celestial.com

Hi there,

this seems to be a problem of the ZyXEL 1496 modem series. I have a 1496E which behaves exactly the
same when communicating with some fax machines. Does anybody know how to fix this without changing 
the modem? Has anybody tried using a newer firmware revision? 

	Alexandre
--
Alexandre Meissonnier
Dohle Handelsgruppe Systemberatung GmbH
53719 Siegburg (Germany)
ameissonnier@dohle.com

-----Ursprngliche Nachricht-----
Von:	Andreas Busse [SMTP:ab@ivm.net]
Gesendet am:	Montag, 6. Oktober 1997 17:34
An:	flexfax@sgi.com
Cc:	Andreas Busse
Betreff:	flexfax: Problems with MPS (more pages, same document)


Hello there,

I'm using the latest HylaFax (hylafax-v4.0pl1) on a Linux 2.0.27 box
together with a Zyxel 1496EG+ and have problems with sending
multi-page documents to some other fax machines. Here's a part
of the debug output:

Oct 06 15:13:06.12: [12876]: SEND begin page
Oct 06 15:13:44.59: [12876]: SENT 46502 bytes of data
Oct 06 15:13:44.59: [12876]: SEND 1D RTC
Oct 06 15:13:44.59: [12876]: SEND end page
Oct 06 15:13:44.59: [12876]: SEND send MPS (more pages, same document)
Oct 06 15:14:00.28: [12876]: --> [5:ERROR]
Oct 06 15:14:00.28: [12876]: SEND recv RTN (retrain negative)
Oct 06 15:14:00.28: [12876]: <-- [7:AT+FDT\r]
Oct 06 15:14:06.11: [12876]: --> [20:+FCS:1,3,0,2,0,0,0,3]
Oct 06 15:14:06.11: [12876]: --> [7:CONNECT]
Oct 06 15:14:06.11: [12876]: SEND begin page
Oct 06 15:14:44.58: [12876]: SENT 46502 bytes of data
Oct 06 15:14:44.58: [12876]: SEND 1D RTC
Oct 06 15:14:44.58: [12876]: SEND end page
Oct 06 15:14:44.58: [12876]: SEND send MPS (more pages, same document)
Oct 06 15:15:00.25: [12876]: --> [5:ERROR]
Oct 06 15:15:00.25: [12876]: SEND recv RTN (retrain negative)

I.e, Hylafax sends the first page, send MPS, gets an error message
back and retransmits the same page, resulting in the same error
again. And so on. As you can imagine, users with such fax machines
complain about getting the same page 9 times (3 times for each trial).
Even worse, the first page is a cover page only, so these users
never see the real message text.

If you have any suggestions, please let me know. I'm not subscribed
to list, so please mail me directly!

With many thanks in advance,
Andy


--
-------------------------------------------------------------- ab@ivm.net ---
  | \    / |\  /|  Gesellschaft fuer   | Andreas Busse
  |  \  /  | \/ |  Internet,           | Im Hufen Boden 16, D-53498 Waldorf
  |   \/   |    |  Vernetzung und      | tel 02636-97690 fax 02636-9769-999
                   Mehrwertdienste mbH | info@ivm.net - http://www.ivm.net/
-----------------------------------------------------------------------------

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 6 Oct 1997 23:31:07 +0200 (MET DST)
From: Thomas Tornblom <Thomas.Tornblom@Hax.SE>
Reply-To: Thomas Tornblom <Thomas.Tornblom@Hax.SE>
Subject: Re: AW: flexfax: Problems with MPS (more pages, same document)
To: ab@ivm.net, ameissonnier@dohle.com
Cc: flexfax@sgi.com, support@zyxel.hinet.net
Sender: owner-flexfax@celestial.com

Add:

Class2DCCQueryCmd:     "!(0,1),(0-5),(0-2),(0-2),(0,1),(0),(0),(0-7)"

to the config file. This disables the ECM support for the modem. I'm not sure if 
ZyXEL ever has gotten around to fix this. I believe the bug is still present, 
both for the 1496 and 2864 series modem.

Care to comment ZyXEL support?

Thomas

> 
> Hi there,
> 
> this seems to be a problem of the ZyXEL 1496 modem series. I have a 1496E which 
behaves exactly the
> same when communicating with some fax machines. Does anybody know how to fix 
this without changing 
> the modem? Has anybody tried using a newer firmware revision? 
> 
> 	Alexandre
> --
> Alexandre Meissonnier
> Dohle Handelsgruppe Systemberatung GmbH
> 53719 Siegburg (Germany)
> ameissonnier@dohle.com
> 
> -----Ursprngliche Nachricht-----
> Von:	Andreas Busse [SMTP:ab@ivm.net]
> Gesendet am:	Montag, 6. Oktober 1997 17:34
> An:	flexfax@sgi.com
> Cc:	Andreas Busse
> Betreff:	flexfax: Problems with MPS (more pages, same document)
> 
> 
> Hello there,
> 
> I'm using the latest HylaFax (hylafax-v4.0pl1) on a Linux 2.0.27 box
> together with a Zyxel 1496EG+ and have problems with sending
> multi-page documents to some other fax machines. Here's a part
> of the debug output:
> 
> Oct 06 15:13:06.12: [12876]: SEND begin page
> Oct 06 15:13:44.59: [12876]: SENT 46502 bytes of data
> Oct 06 15:13:44.59: [12876]: SEND 1D RTC
> Oct 06 15:13:44.59: [12876]: SEND end page
> Oct 06 15:13:44.59: [12876]: SEND send MPS (more pages, same document)
> Oct 06 15:14:00.28: [12876]: --> [5:ERROR]
> Oct 06 15:14:00.28: [12876]: SEND recv RTN (retrain negative)
> Oct 06 15:14:00.28: [12876]: <-- [7:AT+FDT\r]
> Oct 06 15:14:06.11: [12876]: --> [20:+FCS:1,3,0,2,0,0,0,3]
> Oct 06 15:14:06.11: [12876]: --> [7:CONNECT]
> Oct 06 15:14:06.11: [12876]: SEND begin page
> Oct 06 15:14:44.58: [12876]: SENT 46502 bytes of data
> Oct 06 15:14:44.58: [12876]: SEND 1D RTC
> Oct 06 15:14:44.58: [12876]: SEND end page
> Oct 06 15:14:44.58: [12876]: SEND send MPS (more pages, same document)
> Oct 06 15:15:00.25: [12876]: --> [5:ERROR]
> Oct 06 15:15:00.25: [12876]: SEND recv RTN (retrain negative)
> 
> I.e, Hylafax sends the first page, send MPS, gets an error message
> back and retransmits the same page, resulting in the same error
> again. And so on. As you can imagine, users with such fax machines
> complain about getting the same page 9 times (3 times for each trial).
> Even worse, the first page is a cover page only, so these users
> never see the real message text.
> 
> If you have any suggestions, please let me know. I'm not subscribed
> to list, so please mail me directly!
> 
> With many thanks in advance,
> Andy
> 
> 
> --
> -------------------------------------------------------------- ab@ivm.net ---
>   | \    / |\  /|  Gesellschaft fuer   | Andreas Busse
>   |  \  /  | \/ |  Internet,           | Im Hufen Boden 16, D-53498 Waldorf
>   |   \/   |    |  Vernetzung und      | tel 02636-97690 fax 02636-9769-999
>                    Mehrwertdienste mbH | info@ivm.net - http://www.ivm.net/
> -----------------------------------------------------------------------------
> 
> 
> 

Real life:      Thomas Trnblom             Email:  Thomas.Tornblom@Hax.SE
Snail mail:     HB Hax                      Phone:    +46  18 290 290
                Banvallsvgen 14            Fax:      +46  18 290 291
                S - 754 40 Uppsala, Sweden  Cellular: +46 708 323118

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
To: "Alexandre Meissonnier (home)" <ameissonnier@dohle.com>
cc: "'Andreas Busse'" <ab@ivm.net>, "'flexfax@sgi.com'" <flexfax@sgi.com>
Reply-to: Shuvam Misra <shuvam@spacenetindia.com>
From: Shuvam Misra <shuvam@spacenetindia.com>
Subject: Re: AW: flexfax: Problems with MPS (more pages, same document) 
Date: Tue, 07 Oct 1997 05:01:20 +0530
Sender: owner-flexfax@celestial.com

| this seems to be a problem of the ZyXEL 1496 modem series. I have a =
| 1496E which behaves exactly the
| same when communicating with some fax machines. Does anybody know how to =
| fix this without changing=20
| the modem? Has anybody tried using a newer firmware revision?=20

I may be wrong, but I think this is a FAQ. I think the problem lies with
the ECM facility of ZyXEL modems. Switch off the error correction by
putting a line in your config file that tells hylafax not to probe the
modem's error correction facility, but just use it under the assumption
that ECM doesn't work. I remember this being mentioned in the FAQ
somewhere. Please try and see if you can find it there. The exact line
to put into the config files is given there.

Shuvam

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: "Gary Ng (Winclient)" <garyn@winclient.com>
To: flexfax@sgi.com, "'Bruce M. Hahne'" <hahne@aisol.net>
Cc: thomas.tornblom@hax.se, hahne@goemon.aisol.net, support@zyxel.com
Subject: RE: AW: flexfax: Problems with MPS (more pages, same document)
Date: Tue, 7 Oct 1997 11:19:01 +0800
Sender: owner-flexfax@celestial.com

This is because a RTN is sent from the receiving side.
We have seen this before for some fax machines. Ricoh FAX1700L
is one of them. Our work around is to ignore the nack and assume
it is ok !

Gary

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 14 Oct 1997 15:58:46 +0100 (MET)
From: Andreas Busse <ab@ivm.net>
To: "Bruce M. Hahne" <hahne@aisol.net>
cc: flexfax@sgi.com, thomas.tornblom@hax.se, support@zyxel.com
Subject: Re: AW: flexfax: Problems with MPS (more pages, same document)
Sender: owner-flexfax@celestial.com

On Mon, 6 Oct 1997, Bruce M. Hahne wrote:

Hi,

> for several destination fax numbers.  It tends, as one might imagine, to get
> the recipient very upset when they get 9 or so copies of the same first page,
> and nothing else.

in fact -- this is my problem. We're running a TPC.INT cell here, and
as often this problem occurs, people call us and are *very* upset about
abusing their fax machines. Even worse, all they see is the cover sheet,
not the actual message. 

> I gave Zyxel support one of the several destination fax numbers which was
> giving us problems and their response was "problem noted; switch to class 1
> for that destination number".  That's not acceptable; this is an automated
> system and we can't just flip back and forth between class 2.0 and class 1
> on a whim.

Perhaps I shall add a line to the cover page saying: "In case you receive
this page multiple times, blame Zyxel, not us" ;-)

> Our solution was to change modem vendors.  For outbound faxes we now use
> modems with a Lucent chipset in them.  (But I'd switch back if I could get a
> working fix...)

Ok, what type of modem do you use know?
Someone else recommended a Multitech, which I haven't handy right now.
What I have is an old USR Courier V32bis, tons of USR Sportster 28800
and a couple of ELSA/TKR 28800 modems. And finally, a no-name Rockwell
14400 modem with no docs available. And perhaps even more exotic things
somewhere downstairs in dusty corners. 

I swapped to an old 1496E (not EG+ or the like) which at least sometimes
behaves better. I'm not sure about since it passed a couple of test
to a known faxmachine which never worked together with the 1496EG+, 
but it passed not _all_ tests. This is really driving me crazy :-(

Thanks for any help,
Andy

--
-------------------------------------------------------------- ab@ivm.net ---
  | \    / |\  /|  Gesellschaft fuer   | Andreas Busse
  |  \  /  | \/ |  Internet,           | Im Hufen Boden 16, D-53498 Waldorf
  |   \/   |    |  Vernetzung und      | tel 02636-97690 fax 02636-9769-999
                   Mehrwertdienste mbH | info@ivm.net - http://www.ivm.net/
-----------------------------------------------------------------------------

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 15 Oct 1997 08:28:05 +0100
Reply-To: moeller.heinemann@t-online.de
Organization: A. Heinemann Heide GmbH & Co. KG
To: flexfax@sgi.com
Subject: Re: AW: flexfax: Problems with MPS (more pages, same document)
From: moeller.heinemann@t-online.de (Michael Mller)
Sender: owner-flexfax@celestial.com

> in fact -- this is my problem. We're running a TPC.INT cell here, and
> as often this problem occurs, people call us and are *very* upset about
> abusing their fax machines. Even worse, all they see is the cover sheet,
> not the actual message.

I think I've got the same problem here. On some destinations I always
get an
RTN and the page is transmitted nine times (3x3). The workaround I found
is,
when i send the fax in fine mode there are no problems, only the
telephonratings
are higher (twice as normal). I'm now thinking about patching the
HylaFAX-Source
(faxd/Class2Send.c++, removing "morePages = TRUE" in line284). Would
this solve
the problem or will I get other problems ?

Conf:
Linux 2.0.30
Hylafax 4.0pl1
4x Skyconnect 28800

Michael

-- 
Michael Mller                    A. Heinemann Heide GmbH & Co.KG
Unix-Systemadministrator          Technische Gebudeausrstung
moeller.heinemann@t-online.de     Sderdamm 5, 25746 Heide, Germany
D2:0172/4576755  Fax:05334/7485 oder 0481/87654  Tel.:0481/8571-0

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
X-Authentication-Warning: tiehack.gilasoft.com: sfreed owned process doing -bs
Date: Fri, 10 Oct 1997 17:47:29 -0600 (MDT)
From: Steven Freed <sfreed@gilasoft.com>
Reply-To: sfreed@gilasoft.com
To: Gary Ng <garyn@winclient.com>
cc: flexfax@sgi.com
Subject: flexfax: Re: Problems with MPS (more pages, same document)
Sender: owner-flexfax@celestial.com


In your reply to this, you said:

   This is because a RTN is sent from the receiving side.
   We have seen this before for some fax machines. Ricoh FAX1700L
   is one of them. Our work around is to ignore the nack and assume
   it is ok !

What did you do to "ignore the nack"? We are having this same problem, we
have a Zyxel 1496 and it only happens when connecting to a few fax
machines. I have the lastest firmware from Zyxel, and I am pretty sure it
is not an ECM problem since we do have the line

   Class2DCCQueryCmd:     "!(0,1),(0-5),(0-2),(0-2),(0,1),(0),(0),(0-7)"

in our config file and if I take it out we have bad problems to almost
every machine we connect to.

Does anyone out there use a US Robotics Courier V.everything for faxing?
How well does it work? Better than the Zyxels?

Thanks,
Steven.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: dave@delta.demon.co.uk (David Godfrey)
Subject: Re: flexfax: Re: Problems with MPS (more pages, same document)
To: sfreed@gilasoft.com
Date: Sat, 11 Oct 1997 14:35:26 +0100 (BST)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com


> Does anyone out there use a US Robotics Courier V.everything for faxing?
> How well does it work? Better than the Zyxels?

I used a Courier V.everything until about 18 months ago. I had a few
problems with it missing in-coming fax and occasional grief sending,
though I don't remember exactly what. I swapped it with a MultiTech
MT2834ZDX which has worked flawlessly - I recommend it over USR for
faxing.

Regards

-- 
Dave Godfrey
dave@delta.demon.co.uk
Fax +44 181 286 1909
PGP key available

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Sat, 11 Oct 1997 11:37:05 -0700
From: Brian Ng <brian@www.frtnet.com>
Organization: ITIS TECHNOLOGY
To: sfreed@gilasoft.com
CC: flexfax@sgi.com
Subject: Re: flexfax: Re: Problems with MPS (more pages, same document)
Sender: owner-flexfax@celestial.com

Steven Freed wrote:

> What did you do to "ignore the nack"? We are having this same problem, we
> have a Zyxel 1496 and it only happens when connecting to a few fax
> machines. I have the lastest firmware from Zyxel, and I am pretty sure it
> is not an ECM problem since we do have the line
> 
>    Class2DCCQueryCmd:     "!(0,1),(0-5),(0-2),(0-2),(0,1),(0),(0),(0-7)"
> 
> in our config file and if I take it out we have bad problems to almost
> every machine we connect to.
> 
> Does anyone out there use a US Robotics Courier V.everything for faxing?
> How well does it work? Better than the Zyxels?
> 
> Thanks,
> Steven.

	I got exactly the same problem with zyxel 1496e+. It is funny 
that when I use class 2.0, some fax machine does not work. However, 
when I switch to class 2, it works with those machine while another
set of fax machine does not work ( i.e. receving the 1st page 
serval times... then fail).
	Does any one has better recommendation for another
modem ?

	thanks
Brian

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
To: Brian Ng <brian@www.frtnet.com>
cc: sfreed@gilasoft.com, flexfax@sgi.com
Subject: Re: flexfax: Re: Problems with MPS (more pages, same document) 
Date: Sun, 12 Oct 1997 11:21:33 +0100
From: "Mr. Arlington Hewes" <tpcadmin@info.tpc.int>
Sender: owner-flexfax@celestial.com

>>>>> On Sat, 11 Oct 1997, "BN" == Brian Ng wrote:

  BN> I got exactly the same problem with zyxel 1496e+. It is funny that
  BN> when I use class 2.0, some fax machine does not work. However, when I
  BN> switch to class 2, it works with those machine while another set of
  BN> fax machine does not work ( i.e. receving the 1st page serval times...
  BN> then fail).  Does any one has better recommendation for another modem
  BN> ?

Multitech. Have ten, love them all.

;-)

-D

-- 
Mr. Arlington Hewes           (tpcadmin@info.tpc.int)
The TPC.INT Subdomain          (http://www.tpc.int/)

       **************************************************
       ***  FOR GENERAL INFORMATION                   ***
       ***         Mail to tpcfaq@info.tpc.int        ***
       ***  FOR A LIST OF CURRENT COVERAGE            ***
       ***         Mail to tpccover@info.tpc.int      ***
       **************************************************

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Sun, 12 Oct 1997 12:30:15 +0100 (MET)
From: Andreas Busse <ab@ivm.net>
To: Steven Freed <sfreed@gilasoft.com>
cc: Gary Ng <garyn@winclient.com>, flexfax@sgi.com, Andreas Busse <ab@ivm.net>
Subject: Re: flexfax: Re: Problems with MPS (more pages, same document)
Sender: owner-flexfax@celestial.com

On Fri, 10 Oct 1997, Steven Freed wrote:

Hi,

I would also like to know what to do to ignore the NACK. I put in
the Class2DCCQueryCmd line recommended by someone else, but it
doesn't help. Now the remote fax now drops the line. Not better 
at all...

Regards,
Andy

> 
> In your reply to this, you said:
> 
>    This is because a RTN is sent from the receiving side.
>    We have seen this before for some fax machines. Ricoh FAX1700L
>    is one of them. Our work around is to ignore the nack and assume
>    it is ok !
> 
> What did you do to "ignore the nack"? We are having this same problem, we
> have a Zyxel 1496 and it only happens when connecting to a few fax
> machines. I have the lastest firmware from Zyxel, and I am pretty sure it
> is not an ECM problem since we do have the line
> 
>    Class2DCCQueryCmd:     "!(0,1),(0-5),(0-2),(0-2),(0,1),(0),(0),(0-7)"
> 
> in our config file and if I take it out we have bad problems to almost
> every machine we connect to.
> 
> Does anyone out there use a US Robotics Courier V.everything for faxing?
> How well does it work? Better than the Zyxels?
> 
> Thanks,
> Steven.
> 
> 

--
-------------------------------------------------------------- ab@ivm.net ---
  | \    / |\  /|  Gesellschaft fuer   | Andreas Busse
  |  \  /  | \/ |  Internet,           | Im Hufen Boden 16, D-53498 Waldorf
  |   \/   |    |  Vernetzung und      | tel 02636-97690 fax 02636-9769-999
                   Mehrwertdienste mbH | info@ivm.net - http://www.ivm.net/
-----------------------------------------------------------------------------

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Sat, 04 Oct 1997 16:00:32 +0300
From: Uri Teklan <uri@teklan.com.tr>
To: Nico Garcia <raoul@cirl.meei.harvard.edu>,
        "'flexfax@sgi.com'" <flexfax@sgi.com>
CC: Evan Leibovitch <evan@telly.org>
Subject: flexfax: Sending blank pages 
Sender: owner-flexfax@celestial.com

Hi !

We have recently installed hylafax from the binary distribution 
(hylafax-sparc-sunos5.5-v4.0pl1-pkg) on a Sun SPARC 20, Solaris 2.5. 
Our modem is Motorola 28.8. When I try to send a Postscript document
using sendfax, we receive several blank pages on the receiving fax.

Could anyone give some hints? Where should I look in order to solve this
problem?

Thanks

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Sun, 5 Oct 1997 15:29:44 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Uri Teklan <uri@teklan.com.tr>
Cc: "'flexfax@sgi.com'" <flexfax@sgi.com>, Evan Leibovitch <evan@telly.org>
Subject: flexfax: Re: Sending blank pages 
Sender: owner-flexfax@celestial.com

On Sat, 4 Oct 1997, Uri Teklan wrote:

> We have recently installed hylafax from the binary distribution 
> (hylafax-sparc-sunos5.5-v4.0pl1-pkg) on a Sun SPARC 20, Solaris 2.5. 
> Our modem is Motorola 28.8. When I try to send a Postscript document
> using sendfax, we receive several blank pages on the receiving fax.
> 
> Could anyone give some hints? Where should I look in order to solve this
> problem?

Well, first use something to view the Postscript and make sure it is
valid (ghostview and ghostscript are good,
ftp://aeneas.mit.edu/pub/gnu/ and
ftp://ftp.cs.wisc.edu/pub/ghost/current/). Then us ps2fax to turn it
to tiffg3 and use faxview to examine the document, and try sending
*that*.

I always use two modems on the same Sun box to try sending faxes to
and from HylaFAX before putting a new distribution or patch on line.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 6 Oct 1997 11:04:41 -0700 (PDT)
From: "Cheryl L. Huntington" <clh@mellis.com>
To: flexfax@sgi.com, raoul@cirl.meei.harvard.edu, uri@teklan.com.tr
Subject: Re: flexfax: Sending blank pages
Cc: evan@telly.org
Sender: owner-flexfax@celestial.com


I have this problem sending to an HP fax machine. I tried sending to 
another fax machine and it worked fine. I worked hard at trying to 
figure out what the problem was, but never did. I am running on
BSD, HylaFax v4.0p10 with a Zylex modem.

Have you tried sending to other fax machines?

Cheryl


>From owner-flexfax-outbound@celestial.com Sat Oct  4 07:37:46 1997
Date: Sat, 04 Oct 1997 16:00:32 +0300
From: Uri Teklan <uri@teklan.com.tr>
To: Nico Garcia <raoul@cirl.meei.harvard.edu>,
        "'flexfax@sgi.com'" <flexfax@sgi.com>
CC: Evan Leibovitch <evan@telly.org>
Subject: flexfax: Sending blank pages 

Hi !

We have recently installed hylafax from the binary distribution 
(hylafax-sparc-sunos5.5-v4.0pl1-pkg) on a Sun SPARC 20, Solaris 2.5. 
Our modem is Motorola 28.8. When I try to send a Postscript document
using sendfax, we receive several blank pages on the receiving fax.

Could anyone give some hints? Where should I look in order to solve this
problem?

Thanks

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 07 Oct 1997 12:42:55 -0400 (EDT)
Organization: The Creative Group
From: Patrick Gardella <patrick@cre8tivegroup.com>
To: flexfax@celestial.com
Subject: flexfax: Error 425?
Sender: owner-flexfax@celestial.com

I got the following error:

   ----- The following addresses had permanent fatal errors -----
"|/usr/local/bin/faxmail -d -n joe@526-8100"
    (expanded from: <joe@fakename.com>)

   ----- Transcript of session follows -----
faxmail: 425 Cannot create data socket (127.0.0.1,4558): Address already in 
use..
554 "|/usr/local/bin/faxmail -d -n joe6@5226-8100"... unknown mailer
error 255

What does this mean???  And how can I fix it?

I had at most 20 faxes in the queue.  

I am running HylaFAX 4.0pl1 on FreeBSD 2.2.1 with 7 modems.

Patrick

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Comments: Authenticated sender is <jawildman@mail.cfanet.com>
From: "Jim Wildman" <jawildman@cfanet.com>
To: flexfax@sgi.com
Date: Thu, 9 Oct 1997 11:53:15 +0000
Subject: flexfax: Using ModemClass
Reply-to: jawildman@cfanet.com
Sender: owner-flexfax@celestial.com

OS = RH 4.2/2.0.30
HylaFax = Version 4.0pl1
8 port Equinox serial card installed for modems.

3 modems (right now) on cua0, cuq1a1 and cuq1a2
I want to restrict outbound faxing to particular modems.

in etc/config I put
ModemClass:	"out:cuq*"

So how do I tell HylaFax to just use the 'out' group?  Is it automatic?  What 
if I want different users to use different modems?  (ie, I define more 
ModemClass's as I add modems).  
Jim Wildman
Madison County Hospital
210 North Main
London, OH 43140
614/852-1372 x571

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 9 Oct 97 17:12:41 EDT
From: Glenn Burkhardt <glenn@aoi.ultranet.com>
To: flexfax@sgi.com
Subject: flexfax: bug in tkhylafax-3.0 (Patch)
Sender: owner-flexfax@celestial.com

I've found what appears to be another bug - the cover sheet selection
didn't work at all.  Selecting 'No Cover Page' didn't prevent sending a
cover page; selecting any cover page other than the default didn't work
either.

Here's my patch.  It include my previous patch to allow special
characters to appear in the comments section.  

Thanks.


*** orig/faxIt.t	Thu Oct  9 16:58:22 1997
--- faxIt.t	Thu Oct  9 16:59:21 1997
***************
*** 67,76 ****
  
      if {[string length $Th_Info(name)]} {
          # This MUST appear.
!         append options " -d \"$Th_Info(name)@$Th_Info(phone)\""
  
          # Make an ID string for later usage
!         append options " -i \"$Th_Info(name)@$Th_Info(company)\"" 
      } else {
          # This MUST appear.
          append options " -d $Th_Info(phone)"
--- 67,78 ----
  
      if {[string length $Th_Info(name)]} {
          # This MUST appear.
! 	regsub -all {["'$\\]} $Th_Info(name) {\\&} name
!         append options " -d \"$name@$Th_Info(phone)\""
  
          # Make an ID string for later usage
! 	regsub -all {["'$\\]} $Th_Info(company) {\\&} company
!         append options " -i \"$name@$company\"" 
      } else {
          # This MUST appear.
          append options " -d $Th_Info(phone)"
***************
*** 88,115 ****
  	    append faxcmd "[mkCover] | sendfax -n" $options
  	}
      } elseif {$sz > 0} {
  	# Make command. Ignore the following options: -k -s -h -v 
! 	append faxcmd "sendfax "
  
!          foreach ln [lsort [array name Comments]] { 
  	    append comments "$Comments($ln) "
!          }    
  
  	set comments [string trim $comments]
  	if {[string length $comments]} {
  	    append faxcmd " -c \"$comments\""
  	}
  
  	if {[string length $Th_Info(regard)]} {
! 	    append faxcmd " -r \"$Th_Info(regard)\""
  	}
  
  	if {[string length $Th_Info(company)]} {
! 	    append faxcmd " -x \"$Th_Info(company)\""
  	}
  
  	if {[string length $Th_Info(location)]} {
! 	    append faxcmd " -y \"$Th_Info(location)\""
  	}
  
  	append faxcmd $options
--- 90,122 ----
  	    append faxcmd "[mkCover] | sendfax -n" $options
  	}
      } elseif {$sz > 0} {
+ 
  	# Make command. Ignore the following options: -k -s -h -v 
! 	append faxcmd "sendfax " $Th_Info(coverPage)
  
!         foreach ln [lsort [array name Comments]] { 
  	    append comments "$Comments($ln) "
!         }    
  
  	set comments [string trim $comments]
  	if {[string length $comments]} {
+ 	    regsub -all {["'$\\]} $comments {\\&} comments
  	    append faxcmd " -c \"$comments\""
  	}
  
  	if {[string length $Th_Info(regard)]} {
! 	    regsub -all {["'$\\]} $Th_Info(regard) {\\&} regarding
! 	    append faxcmd " -r \"$regarding\""
  	}
  
  	if {[string length $Th_Info(company)]} {
! 	    regsub -all {["'$\\]} $Th_Info(company) {\\&} company
! 	    append faxcmd " -x \"$company\""
  	}
  
  	if {[string length $Th_Info(location)]} {
! 	    regsub -all {["'$\\]} $Th_Info(location) {\\&} location
! 	    append faxcmd " -y \"$location\""
  	}
  
  	append faxcmd $options

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 08 Oct 1997 11:06:47 +0200
From: "Paul P.H. Wilson" <wilson@maple.fzk.de>
Organization: Forschungszentrum Karlrushe, INR
To: HylaFAX Mailing List <flexfax@sgi.com>
Subject: flexfax: Server/Session Logs: Who replies to whom
Sender: owner-flexfax@celestial.com

Hello again,

Yesterday I posted a message about my inability to recieve
faxes.  At that time, I was not sure if my modem was
correctly responding to Class 2, Phase B AT+FDT command.
>From further reading (including the HylaFAX source code) it
is no longer clear to me which device/software should be
generating this command.  The Class2 Send routines of
HylaFAX wait for the same "CONNECT" response as the other
documentation suggests after sending the AT+FDT command.

>From where should this response originate?  The sending
faxmodem?  The receiving faxmodem? The receiving fax
software/firmware?

If it is from the sending faxmodem as I now suspect, it
appears that
a) the sending faxmodem of my previous example is not
responding correctly or
b) there is, in fact, no proper connection being made.

Once again, I believe that (b) is more likely.  What could
cause such a thing?  That is, my faxmodem is answering and
even transmitting identity information (+FCI cards), but
when it comes time for the actual fax transmission,
something is wrong.  Could this be flow control???

--
Paul Wilson                wilson@maple.fzk.de
        http://maple.fzk.de/~wilson/
**********************************************
*       My Canada Includes Quebec            *
* Another Environmentalist for Nuclear Power *
**********************************************

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 9 Oct 1997 16:26:50 -0700 (PDT)
From: Thomas Unger <unger@raindrop.seaslug.org>
To: flexfax@sgi.com
Subject: flexfax: MultiTech Multimodem 5600ZDX and Adaptive answer
Sender: owner-flexfax@celestial.com


OK, 

I'm running a newly installed hylafax v4.0-p11 on FreeBSD 2.2.1.

I am attempting to use a new Multitech k56flex modem (MT5600ZDX).  
Rom level is reported by modem to be "V1.01-K56_DLS".

To summerize, I get everything but adaptive answer to work and I need 
some advice in proceeding with that.


A brief history is that I had hylafax version 3.0something working very
well with an older Multitech 28.8 modem.  I got this new modem and have
been able to get everything but adaptive answer to work.  (data dial out,
fax dial out, and either fax in or data in but not both).

The old modem was a Class2 but this new modem is a Class1.  The old would
do adaptive answer and hylafax would correctly route the incoming call.

The new modem seems to do adaptive answer but something does not work
right.  Since I don't have any fax modem command set documenetation I
don't know who is dropping the ball.

But since adaptive answer seems to be only a Class2 thing I tried setting
up the Class1 adaptive answer scheme:

AdaptiveAnswer:         yes             # enable adaptive answer
AnswerRotary:           "fax data"      # answer for fax, then data
ModemAnswerCmd:         ATH+FCLASS=1;A   # default is to answer as fax
ModemAnswerDataCmd:     ATH+FCLASS=0;A  # hangup and answer as data
Class1RecvIdentTimer:   5000 	        # timeout fax answer in 5 secs

But this does not work either.  Rather, it will correctly answer fax calls
but not data calls.  If I change the sequence to "data" then it will
correctly answer data calls.  This is how I've left it for now...

When I talk to the modem with a terminal program I get the impression
that thing are not working *exactly* right.  I can set it to do adaptive
answer, then test it by calling in with data or with fax.  But the result
strings I see from the modem are into quite what I expect.  Same with the
Class1 answer scheme.  If I dial in with data and tell it to do a FAX
answer (at+fclass1;a) my modem claims it connected.  But it didn't!  I
*think* that my modem is lying (broken) which is why HylaFax is to working
correctly.

So, anyone been successful getting this to work?  Any one else had same
problems?  Any help for me?

Finally, if this modem is just not going to work for what I want what 
new k56flex modem is working well?

Thanks,

                                                      ooooo
 -- Tom Unger                                       oo\    oo
    Wet Weather Consulting                         o   \     o
    (206)325-7403                                 o  B I\V I  o
                                                   o     \   o
                                                    oo    \oo
                                                      ooooo

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 10 Oct 1997 08:33:31 -0700 (PDT)
From: Thomas Unger <unger@raindrop.seaslug.org>
To: Miguel Gualdron <mig@miso.wwa.com>
cc: flexfax@sgi.com
Subject: Re: flexfax: MultiTech Multimodem 5600ZDX and Adaptive answer 
Sender: owner-flexfax@celestial.com


> Have you tried the Class2 adaptive answer scheme in Class1?  Some modems like 
> the USR Sportster are capable of this.

I think that I have.  From the following:

at+fclass=0
OK
at+Fclass=1
OK
at+fclass=2
ERROR
at+fclass=2.0
ERROR
AT+FAA=1
OK


I deduce that my modem only supports class 1 operation.  But ALSO that it
supports some sort of adaptive answer.  

So I tried out the adaptive answer by hand.  From a terminal program I put
the modem in adaptive answer mode then called in with a fax machine and a
data modem.  My modem would respond with "FAX" or "DATA" but it then
seemed irratic about saying "CONNECTED".

Plus, when I tried to configure hylafax to use class1 and AT+FAA=1 it did
not work right.  

Thanks for your responce.

Tom

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 10 Oct 1997 01:52:48 -0500 (CDT)
From: David Nelson <dnelson@psa.pencom.com>
Reply-To: dnelson@pencom.com
To: flexfax@sgi.com
Subject: flexfax: Unspecified Transmit Phase B error
Sender: owner-flexfax@celestial.com


HylaFAX v4.0pl1 under RedHat 4.1 using precompiled rpm from sgi.com; USR
Sportster 56K/X2; Firmware: ???

I tired searching the online HylaFAQ and got "The requested URL
/cgi-bin/search.pl was not found on this server."  I read through some
potential condidates and didn't find anything that matched this. I've
removed the binary I/O as requested and I hope that the included trace is
not inapropriate (13 lines).

After finally getting the modem to dial I am unable to maintain a
connection.  The modem dials, the remote answers, they
handshake/negotiate, then hangup. 

Oct 10 01:21:47.84: [ 1514]: DIAL 5551212
Oct 10 01:21:47.84: [ 1514]: <-- [12:ATDT5551212\r]
Oct 10 01:21:59.68: [ 1514]: --> [4:+FCO]
Oct 10 01:22:02.72: [ 1514]: --> [39:+FNF:^[+FCI:"5555551212"^T+FIS:]
Oct 10 01:22:02.72: [ 1514]: REMOTE NSF "^[+FCI:"5555551212"^T+FIS:"
### 6 Lines of I/O to remote fax ###
### 0 Lines of I/O from remote fax ###
Oct 10 01:22:43.68: [ 1514]: REMOTE HANGUP: Unspecified Transmit Phase B
error (code 20)
Oct 10 01:22:43.69: [ 1514]: SEND FAILED: Unspecified Transmit Phase B 
error
Oct 10 01:22:43.69: [ 1514]: <-- [5:ATH0\r]
Oct 10 01:22:43.92: [ 1514]: --> [4:ATH0]
Oct 10 01:22:43.92: [ 1514]: --> [2:OK]
Oct 10 01:22:43.92: [ 1514]: SESSION END

This is the second fax program that I've tried; hoping that HylaFAX would
better handle the problem.  Is my problem really cruddy hardware or
something else?

Thank you,
David Nelson

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 10 Oct 1997 11:11:49 -0300
From: Julio Cesar Hegedus <Spectroman@planethouse.com.br>
Reply-To: juliochegedus@planethouse.com.br
Organization: Planet House ISP
To: "flexfax@sgi.com" <flexfax@sgi.com>
Subject: [Fwd: flexfax: Unspecified Transmit Phase B error]
Sender: owner-flexfax@celestial.com

Hi,

I Using HylaFax in Linux Slackware, with USR 33.6, and I have the same
problem In this list David tell me about the Class 2.0, that simply doesn't
work with HylaFax.  He tell me to try to configure your modem in Class 1,
this workaround maybe work, but in this same problem, I already change the
class of the modem, but this still not working fine, no problem accur when I
do calls in internal extensions, trough PABX, but when I using external phone
lines, the modem couldn't connect with a fax machine.

Same idea?

dnelson@pencom.com wrote:

> HylaFAX v4.0pl1 under RedHat 4.1 using precompiled rpm from sgi.com; USR
> Sportster 56K/X2; Firmware: ???
>
> I tired searching the online HylaFAQ and got "The requested URL
> /cgi-bin/search.pl was not found on this server."  I read through some
> potential condidates and didn't find anything that matched this. I've
> removed the binary I/O as requested and I hope that the included trace is
> not inapropriate (13 lines).
>
> After finally getting the modem to dial I am unable to maintain a
> connection.  The modem dials, the remote answers, they
> handshake/negotiate, then hangup.
>
> Oct 10 01:21:47.84: [ 1514]: DIAL 5551212
> Oct 10 01:21:47.84: [ 1514]: <-- [12:ATDT5551212\r]
> Oct 10 01:21:59.68: [ 1514]: --> [4:+FCO]
> Oct 10 01:22:02.72: [ 1514]: --> [39:+FNF:^[+FCI:"5555551212"^T+FIS:]
> Oct 10 01:22:02.72: [ 1514]: REMOTE NSF "^[+FCI:"5555551212"^T+FIS:"
> ### 6 Lines of I/O to remote fax ###
> ### 0 Lines of I/O from remote fax ###
> Oct 10 01:22:43.68: [ 1514]: REMOTE HANGUP: Unspecified Transmit Phase B
> error (code 20)
> Oct 10 01:22:43.69: [ 1514]: SEND FAILED: Unspecified Transmit Phase B
> error
> Oct 10 01:22:43.69: [ 1514]: <-- [5:ATH0\r]
> Oct 10 01:22:43.92: [ 1514]: --> [4:ATH0]
> Oct 10 01:22:43.92: [ 1514]: --> [2:OK]
> Oct 10 01:22:43.92: [ 1514]: SESSION END
>
> This is the second fax program that I've tried; hoping that HylaFAX would
> better handle the problem.  Is my problem really cruddy hardware or
> something else?
>
> Thank you,
> David Nelson


From dnelson@pencom.com  Wed Feb 17 13:55:47 1999
Message-ID: <37521@bbs.planethouse.com.br>
From: dnelson@pencom.com
To: Spectroman
Date: 10 Oct 1997 10:10:22 EDT
Subject: flexfax: Unspecified Transmit Phase B error


HylaFAX v4.0pl1 under RedHat 4.1 using precompiled rpm from sgi.com; USR
Sportster 56K/X2; Firmware: ???

I tired searching the online HylaFAQ and got "The requested URL
/cgi-bin/search.pl was not found on this server."  I read through some
potential condidates and didn't find anything that matched this. I've
removed the binary I/O as requested and I hope that the included trace is
not inapropriate (13 lines).

After finally getting the modem to dial I am unable to maintain a
connection.  The modem dials, the remote answers, they
handshake/negotiate, then hangup. 

Oct 10 01:21:47.84: [ 1514]: DIAL 5551212
Oct 10 01:21:47.84: [ 1514]: <-- [12:ATDT5551212\r]
Oct 10 01:21:59.68: [ 1514]: --> [4:+FCO]
Oct 10 01:22:02.72: [ 1514]: --> [39:+FNF:^[+FCI:"5555551212"^T+FIS:]
Oct 10 01:22:02.72: [ 1514]: REMOTE NSF "^[+FCI:"5555551212"^T+FIS:"
### 6 Lines of I/O to remote fax ###
### 0 Lines of I/O from remote fax ###
Oct 10 01:22:43.68: [ 1514]: REMOTE HANGUP: Unspecified Transmit Phase B
error (code 20)
Oct 10 01:22:43.69: [ 1514]: SEND FAILED: Unspecified Transmit Phase B 
error
Oct 10 01:22:43.69: [ 1514]: <-- [5:ATH0\r]
Oct 10 01:22:43.92: [ 1514]: --> [4:ATH0]
Oct 10 01:22:43.92: [ 1514]: --> [2:OK]
Oct 10 01:22:43.92: [ 1514]: SESSION END

This is the second fax program that I've tried; hoping that HylaFAX would
better handle the problem.  Is my problem really cruddy hardware or
something else?

Thank you,
David Nelson

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: Unspecified Transmit Phase B error
To: dnelson@pencom.com
Date: Sat, 11 Oct 1997 13:13:55 +0100 (BST)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> 
> 
> HylaFAX v4.0pl1 under RedHat 4.1 using precompiled rpm from sgi.com; USR
> Sportster 56K/X2; Firmware: ???
> 
> removed the binary I/O as requested and I hope that the included trace is
> not inapropriate (13 lines).

If anything it is not enough.
> Oct 10 01:22:02.72: [ 1514]: --> [39:+FNF:^[+FCI:"5555551212"^T+FIS:]
                                            ^                  ^
Illegal characters.  This is characteristic of earlier Sportster V.34bis
modems in class 2.  I think both the illegal characters should really have
been CR-LF pairs.

You will probably have to run the modem in class 1.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 10 Oct 1997 12:58:16 -0500
From: Lachlan Dunlop <lach@lach.net>
To: flexfax@sgi.com
Subject: flexfax: Trouble starting hylafax
Sender: owner-flexfax@celestial.com

Hello,

Let me start out by saying I'm new to hylafax.  I have started this
project (to implement  a faxserver) about 3 days ago.  I have read and
reread the troubleshooting guide, and searched the mail archives ( I
came up this a match, but no solution).

I have hfaxd configured in my inetd.conf as shown in the troubling
shooting guide.

If I try and start hylafax with "hfaxd -I" I get a getpeername error.

If I try to start hylafax with "haxd -i hylafax", I get cannot bind to
port 4559.

So.... I am posting on this list to maybe see how I am suppose to start
hylafax???

latest release of 4.0pl1, running on linux redhat 4.2

Thanks, and I appologize in advance if I have missed something obvious.

Lach

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 10 Oct 1997 16:40:38 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Lachlan Dunlop <lach@lach.net>
Cc: flexfax@sgi.com
Subject: Re: flexfax: Trouble starting hylafax
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Fri, 10 Oct 1997, Lachlan Dunlop wrote:

> Hello,
> 
> Let me start out by saying I'm new to hylafax.  I have started this
> project (to implement  a faxserver) about 3 days ago.  I have read and
> reread the troubleshooting guide, and searched the mail archives ( I
> came up this a match, but no solution).
> 
> I have hfaxd configured in my inetd.conf as shown in the troubling
> shooting guide.
> 
> If I try and start hylafax with "hfaxd -I" I get a getpeername error.

Using an inetd.conf entry for hfaxd is no longer recommended procedure
(it's not your fault, this has confused a number of people). It is
better to use the startup script in the sources, etc/hylafax, and put
it in the /etc/rc.d directories like the other startup scripts.

If you need a copy of it, I can email you one.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBND6Svz/+ItycgIJRAQEhpAP+Jn1tA3v7vcRnwT24e0M8OOZd/MgAeifz
OKWu+fDWkqsgsS3BGVym4/pYp8f2DkOupfkcsFPckHOi/hqd5V6MbiHHvhlJMkyS
frCrTBNUKFdzyFw6Kn64CDeuRiOWAHmFNE49qJkAt5wnPj4nOcCqQCPSxUKfsjTD
vnWritY5rnc=
=q1Tp
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 14 Oct 1997 20:39:07 -0400 (EDT)
From: Rob Newberry <rob@eats.com>
To: Juergen Schneider <kiss@gds.de>
cc: flexfax@sgi.com
Subject: Re: flexfax: Apple Fax Client
Sender: owner-flexfax@celestial.com


> I know there is an Fax Client for Mac-Maschines. 
> But what is the name of the software and where is it available?

It's called MacFlex, and you can get info at 

	http://www.eats.com/MacFlex/

Rob

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 14 Oct 1997 17:24:38 -0400
From: Ramana Juvvadi <juvvadi@lekha.org>
To: flexfax@sgi.com
Subject: flexfax: New RPMs for hylafax-4.0 and libgr-2.0.14
Sender: owner-flexfax@celestial.com


Welcome to the new rpm of hylafax. One of the motivations for building
this was to learn the capbilities of rpm.  Hylafax is a great package,
but it doesn't work out of the box. It requires an elaborate setup. My
aim in building this package is to reduce some of this complexity for
Redhat Linux. I admit that this release falls short of that goal.
However, I hope the situation improves in future releases.

There is already a linux binary at SGI's site. An rpm for hylafax was
built using this binary by Matthew Rice before.  Some of you who tried
to install the rpm might have experienced problems running
'faxsetup'. This is because the rpm doesn't have all the files. After
encouragement from Matthew, I decided to build a new rpm of hylafax
from the source.

The features of the new rpm are: 

   1.	Redhat Linux 4.2 comes with libgr-2.0.9. However, the libtiff
	that comes in libgr-2.0.9 is older than the version required by
	hylafax. Meanwhile libgr has evolved and the latest version is
	libgr-2.0.13. However,libgr-2.0.13 has some shell scripts missing. I
	added them to make it libgr-2.0.14.

   2.	libgr-2.0.14 subsumes the functionality of another package 
	that comes with Redhat 4.2, netpbm-1mar1994. 

   3.	It comes with rpms of the sources of hylafax and libgr. 

   4.	It uses /usr instead of /usr/local in the tradition of rpm 

   5.	It uses the directories /var/spool/hyla, /usr/lib/hyla instead of
	/var/spool/fax, /usr/lib/fax. The reason is that there are two 
	other fax pages for linux, efax and mgetty+sendfax. I feel it 
	is better to have a unique name like
	hyla than a generic one like fax causing mixup 

   6.	The documentation directory is set to /usr/doc instead of /var/httpd 

   7.	There is an executable called xferstats in hylafax clashing with 
	an xferstats from wu-ftpd. Hence its name is
	changed to xferstat 

For more information look at

	http://www.lekha.org/hylafax
	ftp://ftp.lekha.org/pub

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 15 Oct 1997 16:51:27 -0400
From: "Michael A. Irons" <bb851@lafn.org>
Organization: Yet another Linux machine via LA-FreeNet
To: Ramana Juvvadi <juvvadi@lekha.org>
Cc: flexfax@sgi.com
Subject: Re: flexfax: New RPMs for hylafax-4.0 and libgr-2.0.14
Sender: owner-flexfax@celestial.com

Ramana;

  There is _already_ a RPM for HylaFAX on the HylaFAX source  & binaries
server. Could you please use the same file & directory names that the
offical(?) RPM uses! It gets really confusing when there are packages
with conflicting file & directory names floating around. Please follow
the standard names that are already there.
  Yes, there are other fax programs, but I really don't see why people
would be putting 2 or 3 different fax packages on a system. HylaFAX is a
comprehensive package, more so then the others.
   Thank you however for your time & effort though.
-- 

"By the time they had diminished from 50 to 8, the other dwarves began
to suspect 'Hungry' ..."
                -- Gary Larson, "The Far Side"

		aTdHvAaNnKcSe	>THANKS 'in' advance<

					Michael A. Irons
					mailto:bb851@lafn.org

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 20 Oct 1997 09:26:58 +0100
To: HylaFAX List <flexfax@sgi.com>
From: Phil Watkinson <pkw@elgro.co.uk>
Subject: Re: flexfax: New RPMs for hylafax-4.0 and libgr-2.0.14
Cc: Ramana Juvvadi <juvvadi@lekha.org>
Sender: owner-flexfax@celestial.com


On Tue, 14 Oct 1997, Ramana Juvvadi wrote :
>Welcome to the new rpm of hylafax. One of the motivations for building
>this was to learn the capbilities of rpm.  Hylafax is a great package,
>but it doesn't work out of the box. It requires an elaborate setup. My
>aim in building this package is to reduce some of this complexity for
>Redhat Linux. I admit that this release falls short of that goal.
>However, I hope the situation improves in future releases.

[snip]

Good news. The HylaFAx rpm package for Linux boxes will be very 
popular, and perhaps the most popular of the binary distributions.
In this sense, it will be HylaFAX's 'flagship' product, and so has 
to be right.

With this in mind, could I give some feedback?

>   4.	It uses /usr instead of /usr/local in the tradition of rpm 

Fine, okay with me.

>   5.	It uses the directories /var/spool/hyla, /usr/lib/hyla instead of
>	/var/spool/fax, /usr/lib/fax. The reason is that there are two 
>	other fax pages for linux, efax and mgetty+sendfax. I feel it 
>	is better to have a unique name like
>	hyla than a generic one like fax causing mixup 

Not happy about this, though. I agree with Michael Irons that the potential
mixup from multiple fax programs is outweighed by the confusion amongst
users over the location of the spooling area and config files. Let's stick
with /var/spool/fax and /usr/lib/fax.


>   6.	The documentation directory is set to /usr/doc instead of /var/httpd 

Not a problem.


>   7.	There is an executable called xferstats in hylafax clashing with 
>	an xferstats from wu-ftpd. Hence its name is
>	changed to xferstat 

Hmmm. Can understand, and sympathise with the logic; but I feel this is 
one for Matthias Apitz, the offical maintainer, to decide. How about
'sendstats' as an alternative name? - comparable with sendq and recvq, etc.

Thanks for the work, and I know how long these projects can take, so your
efforts *are* appreciated,

Later,
Phil Watkinson,
Boston, UK.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 20 Oct 1997 08:33:07 -0400
To: flexfax@sgi.com
From: Aidan Van Dyk <aidan@magma.ca>
Subject: Re: flexfax: New RPMs for hylafax-4.0 and libgr-2.0.14
Sender: owner-flexfax@celestial.com

At 09:26 AM 10/20/97 +0100, you wrote:
>
>On Tue, 14 Oct 1997, Ramana Juvvadi wrote :
>>Welcome to the new rpm of hylafax. One of the motivations for building
>>this was to learn the capbilities of rpm.  Hylafax is a great package,
>>but it doesn't work out of the box. It requires an elaborate setup. My
>>aim in building this package is to reduce some of this complexity for
>>Redhat Linux. I admit that this release falls short of that goal.
>>However, I hope the situation improves in future releases.
>
>[snip]
>
>Good news. The HylaFAx rpm package for Linux boxes will be very 
>popular, and perhaps the most popular of the binary distributions.
>In this sense, it will be HylaFAX's 'flagship' product, and so has 
>to be right.
>
>With this in mind, could I give some feedback?
>
>>   4.	It uses /usr instead of /usr/local in the tradition of rpm 
>
>Fine, okay with me.

I'm for /usr/local (anything I add goes in contrib or local - just a patter
of personal prefference I guess)

>
>>   5.	It uses the directories /var/spool/hyla, /usr/lib/hyla instead of
>>	/var/spool/fax, /usr/lib/fax. The reason is that there are two 
>>	other fax pages for linux, efax and mgetty+sendfax. I feel it 
>>	is better to have a unique name like
>>	hyla than a generic one like fax causing mixup 
>
>Not happy about this, though. I agree with Michael Irons that the potential
>mixup from multiple fax programs is outweighed by the confusion amongst
>users over the location of the spooling area and config files. Let's stick
>with /var/spool/fax and /usr/lib/fax.

Agree with this partialy.  /var/spool/fax, but I'm /usr/local/lib/fax

>
>>   6.	The documentation directory is set to /usr/doc instead of /var/httpd 
>
>Not a problem.

Is there any way to get the rpm installer to ask, because every linux
distribution puts web root in different area (And I assume that we want it
in the web root?)

>>   7.	There is an executable called xferstats in hylafax clashing with 
>>	an xferstats from wu-ftpd. Hence its name is
>>	changed to xferstat 
>
>Hmmm. Can understand, and sympathise with the logic; but I feel this is 
>one for Matthias Apitz, the offical maintainer, to decide. How about
>'sendstats' as an alternative name? - comparable with sendq and recvq, etc.

Hmm..   The problem with a clash only comes when you try ot oinstall to
/usr.  wu-ftpd is on most systems, therefore xferstats is in /usr/sbin.  If
hylafax goes in /usr/local, no clash ;-)  Maybe a name change is needed
though, if we want to encourage HylaFAX to be standard with most
distributions (SuSE linux does install it, any other's?)

>Thanks for the work, and I know how long these projects can take, so your
>efforts *are* appreciated,

Mucho appreciated...
Aidan
Aidan Van Dyk                        aidan@magma.ca
---------------------------------------------------------------------
Magma Communications Ltd.            Internet Solutions
Unit 210, 52 Antares Drive              for the Ottawa Area
Nepean, ON, K2E 7Z1  CANADA          http://www.magma.ca/

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 16 Oct 1997 09:11:10 +0100
To: HylaFAX List <flexfax@sgi.com>
From: Phil Watkinson <pkw@elgro.co.uk>
Subject: flexfax: SMS Client available - but not for HylaFAX
Sender: owner-flexfax@celestial.com


Hi, 

A person asked if HylaFAX supported SMS paging  - the answer
is no, but I spotted the following lsm in the c.o.l.a newsgroup 
recently and it maybe a substitute : 

Begin3
Title:          sms_client - send messages to mobiles and pagers
Version:        1.0.0
Entered-date:   15 Sep 1997
Description:    A simple client implementation for the Cellnet GSM Short
                Message Service centre using TAP. Allowing you to send
                short messages to mobile phones and pagers.
Keywords:       TAP modem comms serial network mobile pager paging
Author:         angelo@styx.demon.co.uk (Angelo Masci)
Maintained-by:  angelo@styx.demon.co.uk (Angelo Masci)
Primary-site:   sunsite.unc.edu /pub/Linux/apps/serialcomm/machines
		10k sms_client-1.0.0.tgz
		782 sms_client-1.0.0.lsm
Alternate-site: http://lines0.uwic.ac.uk/~ac0412/sms_client
Original-site:  
Platforms:      Linux, libmodem-1.0.0 package and modem
Copying-policy: GPL
End

Just a thought,
Phil Watkinson,
Boston, UK.

From Matthias.Apitz@SOFTCON.de  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: SMS Client available - but not for HylaFAX
To: pkw@elgro.co.uk (Phil Watkinson)
Date: Thu, 16 Oct 1997 14:52:37 +0200 (MSZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps

Phil Watkinson wrote:
    
    
    Hi, 
    
    A person asked if HylaFAX supported SMS paging  - the answer
    is no, but I spotted the following lsm in the c.o.l.a newsgroup 
    recently and it maybe a substitute : 

HylaFAX v4.0pl1 supports SMS paging via provider using IXO/TAP
on normal modem-dialin lines. Works fine for me with the German
GSM provider "D1-TELEKOM". There are also patches for the UCP
protocol (and maybe as a feature in v4.0pl2).

	matthias
    
    Begin3
    Title:          sms_client - send messages to mobiles and pagers
    Version:        1.0.0
    Entered-date:   15 Sep 1997
    Description:    A simple client implementation for the Cellnet GSM Short
                    Message Service centre using TAP. Allowing you to send
                    short messages to mobile phones and pagers.
    Keywords:       TAP modem comms serial network mobile pager paging
    Author:         angelo@styx.demon.co.uk (Angelo Masci)
    Maintained-by:  angelo@styx.demon.co.uk (Angelo Masci)
    Primary-site:   sunsite.unc.edu /pub/Linux/apps/serialcomm/machines
    		10k sms_client-1.0.0.tgz
    		782 sms_client-1.0.0.lsm
    Alternate-site: http://lines0.uwic.ac.uk/~ac0412/sms_client
    Original-site:  
    Platforms:      Linux, libmodem-1.0.0 package and modem
    Copying-policy: GPL
    End
    
    Just a thought,
    Phil Watkinson,
    Boston, UK.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Sat, 11 Oct 1997 08:18:09 -0500 (CDT)
From: Lachlan Dunlop <lach@lach.net>
To: flexfax@sgi.com
Subject: flexfax: Modem Does not support document size
Sender: owner-flexfax@celestial.com

Hello,

I am getting an error in /var/log/messages.

	"Modem does not support document page width, max page width 2432
pixels.  Image width 1734 pixels."

When I tried a different document to a different fax machine.

	"Modem does not dupport document page width, max page width 1728
pixels.  Image width 1734 pixels."


I am running hylafax 4.0pl1 on Linux RH42, connected to Multitech
1423BA/0110 modem.

Any help would be appreciated.

Thanks

Lach

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Sat, 11 Oct 1997 16:41:26 -0400 (EDT)
From: Dave Wreski <dave@nic.com>
To: Lachlan Dunlop <lach@lach.net>
cc: flexfax@sgi.com
Subject: Re: flexfax: Modem Does not support document size
Sender: owner-flexfax@celestial.com



> 	"Modem does not support document page width, max page width 2432
> pixels.  Image width 1734 pixels."
> 
> When I tried a different document to a different fax machine.

I think this is a client problem.  Check your PS printer driver you are
using in windows, and see if there is a configuration option for 'maximum
compatibility'.  Choose something like the HP 4/PS or Apple PS printer..

Dave

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 16 Oct 1997 09:46:20 +0100
To: HylaFAX List <flexfax@sgi.com>
From: Phil Watkinson <pkw@elgro.co.uk>
Subject: flexfax: Update of html documentation finished
Sender: owner-flexfax@celestial.com


Hi everyone,

The revision of the html documentation ready for the v4.0 patch level 2
has been completed.

They have been uploaded to our website :

	http://www.elgro.demon.co.uk/hylafax/

for testing purposes and everyone is welcome to come and have a look.

My job for this weekend is to create the diff files and email them to
Matthias, so any comments/amendments would be appreciated up to then.

I have been liberal with the material from the mailing-list; those
contributors who can recognize their own words have been mentioned
in the acknowledgements.

Having spent practically all my spare time for two weeks on this project,
I have concluded that my efforts do not come close to Sam's original 
work. But I have tried my best.

Yours,
Phil Watkinson,
Boston, UK.


-------------------------- Details of the Revision ---------------------

*   The docs were organized into chapters. So they have been given  
    numbers. And then re-arranged them to give prominence to the binary
    distributions rather then the sources.

*   Some (but only some) notes have been added to the modems list;
    including notes on the new 56K standards.

*   Added more info on lock files, examples for dialrules & modemclass,
    and considerations for high volume users.

*   Signficantly expanded the mail to fax gateway chapter

*   New troubleshooting topics for faxgetty, email attachments, textfmt
    font problems.

*   Added info about associated programs to the contributed software page.

*   Added the TODO List with other general wishes.

--------------------------------------------------------------------------

From raoul@cirl.meei.harvard.edu  Wed Feb 17 13:55:47 1999
Date: Thu, 16 Oct 1997 13:29:29 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Phil Watkinson <pkw@elgro.co.uk>
Subject: Re: flexfax: Update of html documentation finished

-----BEGIN PGP SIGNED MESSAGE-----

On Thu, 16 Oct 1997, Phil Watkinson wrote:

> Hi everyone,
> 
> The revision of the html documentation ready for the v4.0 patch level 2
> has been completed.
> 
> They have been uploaded to our website :
> 
> 	http://www.elgro.demon.co.uk/hylafax/
> 
> for testing purposes and everyone is welcome to come and have a look.

Good: I'll peruse it this afternoon. If you'd like, you might
nab the PPP information from
	http://cirl.harvard.edu/~raoul/INSTALL-sunos4.1.x-v4.0pl1.html


-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNEZO7j/+ItycgIJRAQE0EgP9HhOlITbw7ZjFw+5hOXx5i2WqDEU4ZHik
COPcjcTYUk+TrbM/al2J4S+OJroTtMliO/uYZcOxGD1g4UcMzo50aLmYZFrwjjyJ
7KTwEkN2lN6iY0RxekP8TCFipnIWPw2UbB0YUwP6DC3JWB0DKFi0iYk36ypPmJHC
dImJk07Zo+4=
=6/rU
-----END PGP SIGNATURE-----

From: Julio C Spinelli <spine001@gold.tc.umn.edu>
To: Phil Watkinson <pkw@elgro.co.uk>
Date: Fri, 17 Oct 1997 21:25:48 +0000
Subject: Re: flexfax: Update of html documentation finished

It would be great if you could add how to use egetty and give a very 
simple example that does nothing but recognized the call as a fax 
call. This will help a lot of people (including me). Same would be 
great for vgetty.

Thanks for the great work,

Julio
> Date:          Thu, 16 Oct 1997 09:46:20 +0100
> To:            HylaFAX List <flexfax@sgi.com>
> From:          Phil Watkinson <pkw@elgro.co.uk>
> Subject:       flexfax: Update of html documentation finished

> 
> Hi everyone,
> 
> The revision of the html documentation ready for the v4.0 patch level 2
> has been completed.
> 
> They have been uploaded to our website :
> 
> 	http://www.elgro.demon.co.uk/hylafax/
> 
> for testing purposes and everyone is welcome to come and have a look.
> 
> My job for this weekend is to create the diff files and email them to
> Matthias, so any comments/amendments would be appreciated up to then.
> 
> I have been liberal with the material from the mailing-list; those
> contributors who can recognize their own words have been mentioned
> in the acknowledgements.
> 
> Having spent practically all my spare time for two weeks on this project,
> I have concluded that my efforts do not come close to Sam's original 
> work. But I have tried my best.
> 
> Yours,
> Phil Watkinson,
> Boston, UK.
> 
> 
> -------------------------- Details of the Revision ---------------------
> 
> *   The docs were organized into chapters. So they have been given  
>     numbers. And then re-arranged them to give prominence to the binary
>     distributions rather then the sources.
> 
> *   Some (but only some) notes have been added to the modems list;
>     including notes on the new 56K standards.
> 
> *   Added more info on lock files, examples for dialrules & modemclass,
>     and considerations for high volume users.
> 
> *   Signficantly expanded the mail to fax gateway chapter
> 
> *   New troubleshooting topics for faxgetty, email attachments, textfmt
>     font problems.
> 
> *   Added info about associated programs to the contributed software page.
> 
> *   Added the TODO List with other general wishes.
> 
> --------------------------------------------------------------------------
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 18 Nov 1997 16:22:01 +0300 (MSK)
To: flexfax@sgi.com
From: sysadm@ns.sirena.ru (Chemisov Sergey)
Subject: flexfax: Where get egetty for FreeBSD
Sender: owner-flexfax@celestial.com

                Hello All !
I trying to use HylaFax v4.0pl1 .
        Faxsetup  tells that it needs  egetty .
Where can I get  egetty  for FreeBSD ?

Chemisov Serg.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Where get egetty for FreeBSD
To: sysadm@ns.sirena.ru (Chemisov Sergey)
Date: Tue, 18 Nov 1997 18:51:09 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Chemisov Sergey wrote:
    
                    Hello All !
    I trying to use HylaFax v4.0pl1 .
            Faxsetup  tells that it needs  egetty .
                            ^^^^^^^^^^^^^
No. It tells that "...The HylaFAX software optionally uses
this program and the fact that it does not exist on the system
is not a fatal error...."

Just ignore this message and read the FAQ, the mailing list
archives and the HTML documentation to learn more about
egetty and vgetty.

    Where can I get  egetty  for FreeBSD ?

If you really need it you'll find some pointers in the
archive how to write it.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Samuel James Liddicott <sam@campbellsci.co.uk>
To: flexfax@sgi.com
MMDF-Warning:  Parse error in original version of preceding line at post.mail.demon.net
Subject: flexfax: Fax != uucp
Date: Wed, 19 Nov 1997 15:04:17 -0000
Sender: owner-flexfax@celestial.com


Well done HylaFax bod! I now got faxing on SCO with an old nasty modem. 

On my system though, I can't set the UID of "fax" to that of "uucp". 

Is this really important?

Thanks

Sam Liddicott                   |   Nothing I say is to be attributed as |
Campbell Scientific Ltd.        | a company statement or representation. |
Campbell Park, 80 Hathern Road, *----------------------------------------+
Shepshed, Leic. United Kingdom. LE12 9AL        Phone: +44 (0) 1509 601141
Email: sam@campbellsci.co.uk                    Fax:   +44 (0) 1509 601091

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Tim Rice <tim@trr.metro.NET>
Subject: Re: flexfax: Fax != uucp
To: sam@campbellsci.co.uk (Samuel James Liddicott)
Date: Wed, 19 Nov 1997 21:24:34 -0800 (PST)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> 
> 
> Well done HylaFax bod! I now got faxing on SCO with an old nasty modem. 
> 
> On my system though, I can't set the UID of "fax" to that of "uucp". 

Why not? It works for me on OS3 and OS5.
....
uucp:x:5:5:UUCP administrator:/usr/lib/uucp:
fax:x:5:5:Facsimile Agent:/usr/local/HylaFAX-v4.0/spool:
....
> 
> Is this really important?

YES!!!

> 
> Thanks
> 
> Sam Liddicott                   |   Nothing I say is to be attributed as |
> Campbell Scientific Ltd.        | a company statement or representation. |
> Campbell Park, 80 Hathern Road, *----------------------------------------+
> Shepshed, Leic. United Kingdom. LE12 9AL        Phone: +44 (0) 1509 601141
> Email: sam@campbellsci.co.uk                    Fax:   +44 (0) 1509 601091
> 
> 

-- 
Tim Rice			Multitalents	(707) 887-1469 (voice)
trr!tim
tim@trr.metro.net
mslbrb!trr!tim@pacbell.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
X-Authentication-Warning: tui.pinnacle.co.nz: jonc owned process doing -bs
Date: Thu, 20 Nov 1997 15:39:33 +1300 (NZDT)
From: Jonathan Chen <jonc@pinnacle.co.nz>
To: Samuel James Liddicott <sam@campbellsci.co.uk>
cc: flexfax@sgi.com
Subject: Re: flexfax: Fax != uucp
Sender: owner-flexfax@celestial.com

On Wed, 19 Nov 1997, Samuel James Liddicott wrote:

> 
> Well done HylaFax bod! I now got faxing on SCO with an old nasty modem. 
> 
> On my system though, I can't set the UID of "fax" to that of "uucp". 
> 
> Is this really important?

Yes, it is. What you've got to do is to twiddle /etc/passwd so that
the uids are the same, and then also look at /tcb/files/auth/f/fax
and alter the u_id# entry so that it matches uucp's (which should be 5
on SCO OS 5.0.x).

Then run `/tcb/bin/authck -av' just to make sure everything's ok with
the tcb subsystem.
--
Jonathan Chen <jonc@pinnacle.co.nz>
----------------------------------------------------------------------
                                        "Nyuck, nyuck, nyuck!" - Curly

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Reply-To: hazard@vicor-nb.com
From: John Hazard <hazard@vicor-nb.com>
To: flexfax@sgi.com
Subject: flexfax: Unspecified Transmit Phase B error   
Date: Thu, 16 Oct 97 11:21:33 PDT
Sender: owner-flexfax@celestial.com



I am having problems with HylaFax on a freebsd machine.  As you can
see below it connects, but quits pretty quickly.  

I have tried to 2 "good" fax machines.  The modem handles data ok,
i.e. I can dial an ISP.  The fax  modem is also OK under Windows 95.


The modem config file is at the end, but here is the head of that file:

#
# Configuration for using the Class 2.0 command interface
# with a USR v.34 Courier or Sporster modem.
#
# RTS/CTS flow control is used default.
#
#

Your facsimile job to 5551616 failed because:
    Unspecified Transmit Phase B error; too many attempts to send

    ---- Transcript of session follows ----

Oct 16 09:19:06.62: [13816]: SESSION BEGIN 00000007 15105551616
Oct 16 09:19:06.62: [13816]: SEND FAX: JOB 2 DEST 5551616 COMMID 00000007
Oct 16 09:19:06.62: [13816]: DELAY 2600 ms
Oct 16 09:19:09.23: [13816]: <-- [17:ATE0V1Q0X4S0=0H0\r]
Oct 16 09:19:09.37: [13816]: --> [2:OK]
Oct 16 09:19:09.37: [13816]: <-- [32:ATS8=2S7=60&H1&I0&R2S13=1&D2&C1\r]
Oct 16 09:19:09.51: [13816]: --> [2:OK]
Oct 16 09:19:09.51: [13816]: <-- [14:AT+FCLASS=2.0\r]
Oct 16 09:19:09.64: [13816]: --> [2:OK]
Oct 16 09:19:09.64: [13816]: <-- [9:AT+FLO=2\r]
Oct 16 09:19:09.77: [13816]: --> [2:OK]
Oct 16 09:19:09.77: [13816]: <-- [9:AT+FPP=0\r]
Oct 16 09:19:09.90: [13816]: --> [2:OK]
Oct 16 09:19:09.90: [13816]: <-- [9:AT+FBO=0\r]
Oct 16 09:19:10.03: [13816]: --> [2:OK]
Oct 16 09:19:10.03: [13816]: <-- [10:AT+FCT=30\r]
Oct 16 09:19:10.17: [13816]: --> [2:OK]
Oct 16 09:19:10.17: [13816]: <-- [15:AT+FNR=1,1,1,1\r]
Oct 16 09:19:10.31: [13816]: --> [2:OK]
Oct 16 09:19:10.31: [13816]: <-- [9:AT+FIE=0\r]
Oct 16 09:19:10.44: [13816]: --> [2:OK]
Oct 16 09:19:10.44: [13816]: <-- [23:AT+FCC=1,5,2,2,0,0,0,0\r]
Oct 16 09:19:10.58: [13816]: --> [2:OK]
Oct 16 09:19:10.58: [13816]: <-- [5:ATM0\r]
Oct 16 09:19:10.71: [13816]: --> [2:OK]
Oct 16 09:19:10.72: [13816]: <-- [14:AT+FCLASS=2.0\r]
Oct 16 09:19:10.96: [13816]: --> [2:OK]
Oct 16 09:19:10.96: [13816]: <-- [9:AT+FLO=2\r]
Oct 16 09:19:11.19: [13816]: --> [2:OK]
Oct 16 09:19:11.19: [13816]: <-- [9:AT+FPP=0\r]
Oct 16 09:19:11.42: [13816]: --> [2:OK]
Oct 16 09:19:11.42: [13816]: <-- [9:AT+FBO=0\r]
Oct 16 09:19:11.65: [13816]: --> [2:OK]
Oct 16 09:19:11.65: [13816]: <-- [10:AT+FCT=30\r]
Oct 16 09:19:11.89: [13816]: --> [2:OK]
Oct 16 09:19:11.89: [13816]: <-- [15:AT+FNR=1,1,1,1\r]
Oct 16 09:19:12.13: [13816]: --> [2:OK]
Oct 16 09:19:12.13: [13816]: <-- [9:AT+FIE=0\r]
Oct 16 09:19:12.36: [13816]: --> [2:OK]
Oct 16 09:19:12.36: [13816]: <-- [23:AT+FCC=1,5,2,2,0,0,0,0\r]
Oct 16 09:19:12.60: [13816]: --> [2:OK]
Oct 16 09:19:12.60: [13816]: <-- [16:AT+FLI="Hazard"\r]
Oct 16 09:19:12.84: [13816]: --> [2:OK]
Oct 16 09:19:12.84: [13816]: DIAL 5551616
Oct 16 09:19:12.84: [13816]: <-- [12:ATDT5551616\r]
Oct 16 09:19:22.08: [13816]: --> [4:+FCO]
Oct 16 09:19:24.54: [13816]: --> [15:+FNF:+FCI:"   ]
Oct 16 09:19:24.54: [13816]: REMOTE NSF "+FCI:""
Oct 16 09:19:24.54: [13816]: --> [32:      5105551616"+FIS:1,5,0,2,1]
Oct 16 09:19:24.54: [13816]: --> [43:0,0,3;<~[<Y\=|i]
Oct 16 09:19:24.65: [13816]: --> [57:@)eQ+v@hmBzdBeB0F`kG{@&]
Oct 16 09:19:25.33: [13816]: --> [18:K]
Oct 16 09:19:25.33: [13816]: --> [58::1,5,0,2,1,0,0,3;<~[<Y\=|i8m]
Oct 16 09:19:25.48: [13816]: --> [25:A"# A}]
Oct 16 09:19:25.95: [13816]: --> [33:tC'lAfqQK@A9]
Oct 16 09:19:26.67: [13816]: --> [2:]
Oct 16 09:19:26.67: [13816]: --> [33:@iWQFOiC?g5v^]
Oct 16 09:19:27.18: [13816]: --> [23: T.s024]
Oct 16 09:19:27.18: [13816]: --> [2:OK]
Oct 16 09:19:27.18: [13816]: REMOTE HANGUP: Unspecified Transmit Phase B error (code 20)
Oct 16 09:19:27.18: [13816]: SEND FAILED: Unspecified Transmit Phase B error
Oct 16 09:19:27.18: [13816]: SEND FAILED: Unspecified Transmit Phase B error; too many attempts to send
Oct 16 09:19:27.18: [13816]: <-- [5:ATH0\r]
Oct 16 09:19:27.41: [13816]: --> [2:OK]
Oct 16 09:19:27.43: [13816]: SESSION END

    ---- Unsent job status ----

   Destination: 5551616
         JobID: 2
       GroupID: 2
        Sender: John Hazard
      Mailaddr: hazard@hazard
        CommID: 00000007
Submitted From: localhost
    Page Width: 215 (mm)
   Page Length: 279 (mm)
    Resolution: 98 (lpi)
        Status: Unspecified Transmit Phase B error; too many attempts to send
       Dialogs: 3 (exchanges with remote device)
         Dials: 0 (consecutive failed calls to destination)
         Calls: 3 (total phone calls placed)
         Pages: 0 (pages transmitted)
      TotPages: 1 (total pages to transmit)
      Attempts: 0 (attempts to send current page)
        Dirnum: 0 (directory of next page to send)

    ---- Documents submitted for transmission ----

The following documents were submitted for transmission and are
available on the server for reuse until they are automatically
purged when this job is removed.  Documents may also be manually
removed using the faxrm command; consult faxrm(1) for information.

Filename                 Size Type
docq/doc2.cover         12519 PostScript cover page
docq/doc3.ps             5752 PostScript


# Configuration for using the Class 2.0 command interface
# with a USR v.34 Courier or Sporster modem.
#
# RTS/CTS flow control is used default.
#
#
CountryCode:		1
AreaCode:		510
FAXNumber:		1-510-555-6239
LongDistancePrefix:	1
InternationalPrefix:	011
DialStringRules:	"etc/dialrules"
ServerTracing:		1
SessionTracing:		11
RecvFileMode:		0666
LogFileMode:		0666
DeviceMode:		0660
RingsBeforeAnswer:	1
SpeakerVolume:		medium
GettyArgs:		"std.%s"
LocalIdentifier:	Hazard
TagLineFont:		etc/lutRS18.pcf
TagLineFormat:		"From %%l|%c|Page %%p of %%t"
MaxRecvPages:		25
#
#
# Modem-related stuff: should reflect modem command interface
# and hardware connection/cabling (e.g. flow control).
#
ModemType:		Class2.0	# use class 2.0 interface
ModemRate:		38400		# DCE-DTE communication rate
ModemFlowControl:	rtscts		# XON/XOFF flow control assumed
#
ModemNoFlowCmd:		AT&H0&I0&R1	# setup modem for no flow control
ModemHardFlowCmd:	AT&H1&I0&R2	# setup modem for hardware flow control
ModemSoftFlowCmd:	AT&H2&I2&R1	# setup modem for software flow control
#
ModemSetupDTRCmd:	ATS13=1&D2	# setup so DTR drop resets modem
ModemSetupDCDCmd:	AT&C1		# setup so DCD reflects carrier (or not)
ModemResultCodesCmd:	ATQ0X4		# enable extended result codes
#
# NB: adaptive answer only seems to work properly when
#     the modem is left idling in Class 2.0
#
ModemSetupAACmd:	AT+FAA=1
#
# Set modem speaker volume commands: OFF QUIET LOW MEDIUM HIGH.
# Note that we both turn the speaker on/off and set volume.
#
ModemSetVolumeCmd:	"ATM0 ATM1 ATM1 ATM1 ATM2"
#
# Modem does not support HDLC frame tracing; we add this just
# to eliminate spurious ERROR results that confuse the naive.
#
Class2BUGCmd:		AT+FBU=0
#
# The modem doesn't support copy quality checking, even though it
# returns (0-2,0-2) for AT+FCQ=?; therefore we override the query
# response so that the server will do copy quality checking.
#
Class2CQQueryCmd:	!(0),(0)	# override modem response

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 16 Oct 1997 19:07:51 -0700 (PDT)
From: John Hazard <hazard@vicor-nb.com>
To: flexfax@sgi.com
Subject: Re: flexfax: Unspecified Transmit Phase B error   
Reply-To: hazard@vicor-nb.com
Sender: owner-flexfax@celestial.com

I guees I'll answer my own question the following line in
cuaa0/config:



Class2NRCmd:            AT+FNR=1,1,1,0  # setup status reporting command


JH

John Hazard writes:
 > 
 > 
 > I am having problems with HylaFax on a freebsd machine.  As you can
 > see below it connects, but quits pretty quickly.  
 > 
 > I have tried to 2 "good" fax machines.  The modem handles data ok,
 > i.e. I can dial an ISP.  The fax  modem is also OK under Windows 95.
 > 
 > 
 > The modem config file is at the end, but here is the head of that file:
 > 
 > #
 > # Configuration for using the Class 2.0 command interface
 > # with a USR v.34 Courier or Sporster modem.
 > #
 > # RTS/CTS flow control is used default.
 > #
 > #
 > 
 > Your facsimile job to 5551616 failed because:
 >     Unspecified Transmit Phase B error; too many attempts to send
 > 
 >     ---- Transcript of session follows ----
 > 
 > Oct 16 09:19:06.62: [13816]: SESSION BEGIN 00000007 15105551616
 > Oct 16 09:19:06.62: [13816]: SEND FAX: JOB 2 DEST 5551616 COMMID 00000007
 > Oct 16 09:19:06.62: [13816]: DELAY 2600 ms
 > Oct 16 09:19:09.23: [13816]: <-- [17:ATE0V1Q0X4S0=0H0\r]
 > Oct 16 09:19:09.37: [13816]: --> [2:OK]
 > Oct 16 09:19:09.37: [13816]: <-- [32:ATS8=2S7=60&H1&I0&R2S13=1&D2&C1\r]
 > Oct 16 09:19:09.51: [13816]: --> [2:OK]
 > Oct 16 09:19:09.51: [13816]: <-- [14:AT+FCLASS=2.0\r]
 > Oct 16 09:19:09.64: [13816]: --> [2:OK]
 > Oct 16 09:19:09.64: [13816]: <-- [9:AT+FLO=2\r]
 > Oct 16 09:19:09.77: [13816]: --> [2:OK]
 > Oct 16 09:19:09.77: [13816]: <-- [9:AT+FPP=0\r]
 > Oct 16 09:19:09.90: [13816]: --> [2:OK]
 > Oct 16 09:19:09.90: [13816]: <-- [9:AT+FBO=0\r]
 > Oct 16 09:19:10.03: [13816]: --> [2:OK]
 > Oct 16 09:19:10.03: [13816]: <-- [10:AT+FCT=30\r]
 > Oct 16 09:19:10.17: [13816]: --> [2:OK]
 > Oct 16 09:19:10.17: [13816]: <-- [15:AT+FNR=1,1,1,1\r]
 > Oct 16 09:19:10.31: [13816]: --> [2:OK]
 > Oct 16 09:19:10.31: [13816]: <-- [9:AT+FIE=0\r]
 > Oct 16 09:19:10.44: [13816]: --> [2:OK]
 > Oct 16 09:19:10.44: [13816]: <-- [23:AT+FCC=1,5,2,2,0,0,0,0\r]
 > Oct 16 09:19:10.58: [13816]: --> [2:OK]
 > Oct 16 09:19:10.58: [13816]: <-- [5:ATM0\r]
 > Oct 16 09:19:10.71: [13816]: --> [2:OK]
 > Oct 16 09:19:10.72: [13816]: <-- [14:AT+FCLASS=2.0\r]
 > Oct 16 09:19:10.96: [13816]: --> [2:OK]
 > Oct 16 09:19:10.96: [13816]: <-- [9:AT+FLO=2\r]
 > Oct 16 09:19:11.19: [13816]: --> [2:OK]
 > Oct 16 09:19:11.19: [13816]: <-- [9:AT+FPP=0\r]
 > Oct 16 09:19:11.42: [13816]: --> [2:OK]
 > Oct 16 09:19:11.42: [13816]: <-- [9:AT+FBO=0\r]
 > Oct 16 09:19:11.65: [13816]: --> [2:OK]
 > Oct 16 09:19:11.65: [13816]: <-- [10:AT+FCT=30\r]
 > Oct 16 09:19:11.89: [13816]: --> [2:OK]
 > Oct 16 09:19:11.89: [13816]: <-- [15:AT+FNR=1,1,1,1\r]
 > Oct 16 09:19:12.13: [13816]: --> [2:OK]
 > Oct 16 09:19:12.13: [13816]: <-- [9:AT+FIE=0\r]
 > Oct 16 09:19:12.36: [13816]: --> [2:OK]
 > Oct 16 09:19:12.36: [13816]: <-- [23:AT+FCC=1,5,2,2,0,0,0,0\r]
 > Oct 16 09:19:12.60: [13816]: --> [2:OK]
 > Oct 16 09:19:12.60: [13816]: <-- [16:AT+FLI="Hazard"\r]
 > Oct 16 09:19:12.84: [13816]: --> [2:OK]
 > Oct 16 09:19:12.84: [13816]: DIAL 5551616
 > Oct 16 09:19:12.84: [13816]: <-- [12:ATDT5551616\r]
 > Oct 16 09:19:22.08: [13816]: --> [4:+FCO]
 > Oct 16 09:19:24.54: [13816]: --> [15:+FNF:+FCI:"   ]
 > Oct 16 09:19:24.54: [13816]: REMOTE NSF "+FCI:""
 > Oct 16 09:19:24.54: [13816]: --> [32:      5105551616"+FIS:1,5,0,2,1]
 > Oct 16 09:19:24.54: [13816]: --> [43:0,0,3;<~[<Y\=|i]
 > Oct 16 09:19:24.65: [13816]: --> [57:@)eQ+v@hmBzdBeB0F`kG{@&]
 > Oct 16 09:19:25.33: [13816]: --> [18:K]
 > Oct 16 09:19:25.33: [13816]: --> [58::1,5,0,2,1,0,0,3;<~[<Y\=|i8m]
 > Oct 16 09:19:25.48: [13816]: --> [25:A"# A}]
 > Oct 16 09:19:25.95: [13816]: --> [33:tC'lAfqQK@A9]
 > Oct 16 09:19:26.67: [13816]: --> [2:]
 > Oct 16 09:19:26.67: [13816]: --> [33:@iWQFOiC?g5v^]
 > Oct 16 09:19:27.18: [13816]: --> [23: T.s024]
 > Oct 16 09:19:27.18: [13816]: --> [2:OK]
 > Oct 16 09:19:27.18: [13816]: REMOTE HANGUP: Unspecified Transmit Phase B error (code 20)
 > Oct 16 09:19:27.18: [13816]: SEND FAILED: Unspecified Transmit Phase B error
 > Oct 16 09:19:27.18: [13816]: SEND FAILED: Unspecified Transmit Phase B error; too many attempts to send
 > Oct 16 09:19:27.18: [13816]: <-- [5:ATH0\r]
 > Oct 16 09:19:27.41: [13816]: --> [2:OK]
 > Oct 16 09:19:27.43: [13816]: SESSION END
 > 
 >     ---- Unsent job status ----
 > 
 >    Destination: 5551616
 >          JobID: 2
 >        GroupID: 2
 >         Sender: John Hazard
 >       Mailaddr: hazard@hazard
 >         CommID: 00000007
 > Submitted From: localhost
 >     Page Width: 215 (mm)
 >    Page Length: 279 (mm)
 >     Resolution: 98 (lpi)
 >         Status: Unspecified Transmit Phase B error; too many attempts to send
 >        Dialogs: 3 (exchanges with remote device)
 >          Dials: 0 (consecutive failed calls to destination)
 >          Calls: 3 (total phone calls placed)
 >          Pages: 0 (pages transmitted)
 >       TotPages: 1 (total pages to transmit)
 >       Attempts: 0 (attempts to send current page)
 >         Dirnum: 0 (directory of next page to send)
 > 
 >     ---- Documents submitted for transmission ----
 > 
 > The following documents were submitted for transmission and are
 > available on the server for reuse until they are automatically
 > purged when this job is removed.  Documents may also be manually
 > removed using the faxrm command; consult faxrm(1) for information.
 > 
 > Filename                 Size Type
 > docq/doc2.cover         12519 PostScript cover page
 > docq/doc3.ps             5752 PostScript
 > 
 > 
 > # Configuration for using the Class 2.0 command interface
 > # with a USR v.34 Courier or Sporster modem.
 > #
 > # RTS/CTS flow control is used default.
 > #
 > #
 > CountryCode:		1
 > AreaCode:		510
 > FAXNumber:		1-510-555-6239
 > LongDistancePrefix:	1
 > InternationalPrefix:	011
 > DialStringRules:	"etc/dialrules"
 > ServerTracing:		1
 > SessionTracing:		11
 > RecvFileMode:		0666
 > LogFileMode:		0666
 > DeviceMode:		0660
 > RingsBeforeAnswer:	1
 > SpeakerVolume:		medium
 > GettyArgs:		"std.%s"
 > LocalIdentifier:	Hazard
 > TagLineFont:		etc/lutRS18.pcf
 > TagLineFormat:		"From %%l|%c|Page %%p of %%t"
 > MaxRecvPages:		25
 > #
 > #
 > # Modem-related stuff: should reflect modem command interface
 > # and hardware connection/cabling (e.g. flow control).
 > #
 > ModemType:		Class2.0	# use class 2.0 interface
 > ModemRate:		38400		# DCE-DTE communication rate
 > ModemFlowControl:	rtscts		# XON/XOFF flow control assumed
 > #
 > ModemNoFlowCmd:		AT&H0&I0&R1	# setup modem for no flow control
 > ModemHardFlowCmd:	AT&H1&I0&R2	# setup modem for hardware flow control
 > ModemSoftFlowCmd:	AT&H2&I2&R1	# setup modem for software flow control
 > #
 > ModemSetupDTRCmd:	ATS13=1&D2	# setup so DTR drop resets modem
 > ModemSetupDCDCmd:	AT&C1		# setup so DCD reflects carrier (or not)
 > ModemResultCodesCmd:	ATQ0X4		# enable extended result codes
 > #
 > # NB: adaptive answer only seems to work properly when
 > #     the modem is left idling in Class 2.0
 > #
 > ModemSetupAACmd:	AT+FAA=1
 > #
 > # Set modem speaker volume commands: OFF QUIET LOW MEDIUM HIGH.
 > # Note that we both turn the speaker on/off and set volume.
 > #
 > ModemSetVolumeCmd:	"ATM0 ATM1 ATM1 ATM1 ATM2"
 > #
 > # Modem does not support HDLC frame tracing; we add this just
 > # to eliminate spurious ERROR results that confuse the naive.
 > #
 > Class2BUGCmd:		AT+FBU=0
 > #
 > # The modem doesn't support copy quality checking, even though it
 > # returns (0-2,0-2) for AT+FCQ=?; therefore we override the query
 > # response so that the server will do copy quality checking.
 > #
 > Class2CQQueryCmd:	!(0),(0)	# override modem response

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: Unspecified Transmit Phase B error
To: hazard@vicor-nb.com
Date: Fri, 17 Oct 1997 08:03:09 +0100 (BST)
Cc: flexfax@celestial.com
Sender: owner-flexfax@celestial.com

> # Configuration for using the Class 2.0 command interface
> # with a USR v.34 Courier or Sporster modem.

> Oct 16 09:19:24.54: [13816]: --> [32:      5105551616"+FIS:1,5,0,2,1]
                                                         ^

The well documented Sportster V.34 class 2 bug.

Revert to class 1, or get a modem that works (supposedly some firmware
version of the Sportster work).

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 17 Oct 1997 15:28:04 +0100
To: HylaFAX List <flexfax@sgi.com>
From: Phil Watkinson <pkw@elgro.co.uk>
Subject: Re: flexfax: Unspecified Transmit Phase B error   
Cc: John Hazard <hazard@vicor-nb.com>
Sender: owner-flexfax@celestial.com

At 19:07 16/10/1997 -0700, John Hazard wrote:
>I guees I'll answer my own question the following line in
>cuaa0/config:
>
>
>
>Class2NRCmd:            AT+FNR=1,1,1,0  # setup status reporting command
>
>
>JH
>

This fix was also suggested by, I believe, Dave Wreski; but unfortunately
in neither case was the actual model of the US Robotics modem actually
mentioned.

Could you let us know?

HylaFAX puts the relevant info in the syslog/messages - for instance mine
says :-

        MODEM USROBOTICS SPORTSTER 28800 FAX
        Product Type            UK External
        Options                 V32bis, V.FC, V.34
        Fax Options             Class 1 / Class 2.0
        Clock Freq              20.16 MHz
        Eprom                   256K
        Ram                     32K
        Supervisor Date         04/28/95
        DSP Date                03/31/95
        Supervisor rev          044-6.0
        DSP rev                 11.3

which, touch wood, works perfectly in a send-only environment.
Zdravko Balorda on the 29 May 97 mentioned that, in his opinion,
anything newer than this :

        Supervisor date        04/18/95
        DSP date               03/31/95
        Supervisor rev         6.0.5
        DSP rev                1.0.9

had problems in Class 2.0 mode. (Perhaps mine just snuck through).

Thanks,
Phil Watkinson,
Boston, UK.

From hazard@vicor-nb.com  Wed Feb 17 13:55:47 1999
Date: Fri, 17 Oct 1997 09:10:21 -0700 (PDT)
From: John Hazard <hazard@vicor-nb.com>
To: Phil Watkinson <pkw@elgro.co.uk>
Subject: Re: flexfax: Unspecified Transmit Phase B error   
Reply-To: hazard@vicor-nb.com

Phil Watkinson writes:
 > This fix was also suggested by, I believe, Dave Wreski; but unfortunately
 > in neither case was the actual model of the US Robotics modem actually
 > mentioned.
 > 
 > Could you let us know?


Yes, I beleive it was Dave who mentioned this, I searched through the
mail list archives.

I could not find the message you mentioned, because i dont have hyla
fax logging, yet, so I just queried the modem:

Product type           US/Canada Internal
Options                V32bis,V.FC,V.34+
Fax Options            Class 1/Class 2.0
Clock Freq             92.0Mhz
Line Options           Caller ID,Distinctive Ring
Voice Options          Speakerphone,TAD
Eprom                  256k
Ram                    64k

EPROM date             4/24/96
DSP date               4/24/96

EPROM rev              2.0
DSP rev                2.0


 > 
 > HylaFAX puts the relevant info in the syslog/messages - for instance mine
 > says :-
 > 
 >         MODEM USROBOTICS SPORTSTER 28800 FAX
 >         Product Type            UK External
 >         Options                 V32bis, V.FC, V.34
 >         Fax Options             Class 1 / Class 2.0
 >         Clock Freq              20.16 MHz
 >         Eprom                   256K
 >         Ram                     32K
 >         Supervisor Date         04/28/95
 >         DSP Date                03/31/95
 >         Supervisor rev          044-6.0
 >         DSP rev                 11.3
 > 
 > which, touch wood, works perfectly in a send-only environment.
 > Zdravko Balorda on the 29 May 97 mentioned that, in his opinion,
 > anything newer than this :
 > 
 >         Supervisor date        04/18/95
 >         DSP date               03/31/95
 >         Supervisor rev         6.0.5
 >         DSP rev                1.0.9
 > 
 > had problems in Class 2.0 mode. (Perhaps mine just snuck through).
 > 
 > Thanks,
 > Phil Watkinson,
 > Boston, UK.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 15 Oct 1997 11:27:13 -0500
From: Lachlan Dunlop <lach@lach.net>
To: flexfax@sgi.com
Subject: flexfax: Ghostscript problem???
Sender: owner-flexfax@celestial.com

Well now I'm going backwards.

I am now getting a Ghostscript error:

    gs: Cannot open X display `(null)'.
    /bin/cat: write error: Broken pipe
    bin/ps2fax: -sDEVICE=tiffg32d: command not found

When I do a gs -?  the tiffg32d is in the device list.

I am running linux RH42 with ghostscript 3.33, is it worth upgrading to
5.0.... or ghostscript???

Lach

PS I know this isn't a ghostscript list, but I thought I could get some help
here.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Alexandre Meissonnier <ameissonnier@dohle.com>
Reply-To: "ameissonnier@dohle.com" <ameissonnier@dohle.com>
To: "'Lachlan Dunlop'" <lach@lach.net>, "flexfax@sgi.com" <flexfax@sgi.com>
Subject: RE: flexfax: Ghostscript problem???
Date: Thu, 16 Oct 1997 07:10:19 +0200
Organization: Dohle Handelsgruppe Systemberatung GmbH
Sender: owner-flexfax@celestial.com



-----Original Message-----
From:	Lachlan Dunlop [SMTP:lach@lach.net]
Sent:	Wednesday, October 15, 1997 6:27 PM
To:	flexfax@sgi.com
Subject:	flexfax: Ghostscript problem???

> I am now getting a Ghostscript error:
> 
>     gs: Cannot open X display `(null)'.
>     /bin/cat: write error: Broken pipe
>     bin/ps2fax: -sDEVICE=tiffg32d: command not found

Did you edit the ps2fax script? I guess that you accidentally deleted the trailing backslash in 
the line just before '-sDEVICE...'.
The shell tries to evaluate this line as a separate command, although it should be treated as a 
part of the gs command.


Hope this helps

	Alexandre
--
-------------------------------------------------------------------------------
Alexandre Meissonnier
Dohle Handelsgruppe Systemberatung GmbH
53719 Siegburg (Germany)
ameissonnier@dohle.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 16 Oct 1997 13:24:00 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Lachlan Dunlop <lach@lach.net>
Cc: flexfax@fddi-sgi.engr.sgi.com
Subject: Re: flexfax: Ghostscript problem???
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Wed, 15 Oct 1997, Lachlan Dunlop wrote:

> Well now I'm going backwards.
> 
> I am now getting a Ghostscript error:
> 
>     gs: Cannot open X display `(null)'.
>     /bin/cat: write error: Broken pipe
>     bin/ps2fax: -sDEVICE=tiffg32d: command not found
> 
> When I do a gs -?  the tiffg32d is in the device list.
> 
> I am running linux RH42 with ghostscript 3.33, is it worth upgrading to
> 5.0.... or ghostscript???

What? Run the ps2fax script by hand with /bin/sh -x, and see what you get.

You also probably want the most recent version of ghostscript, in order
to handle more recent Postscript commands.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNEZNqT/+ItycgIJRAQFDTAP+OycTJPZwEjw9gxaOdozuFGkslNn19Qd0
UqaJq8zynXv+G7fwJ955zRmSQwxpqS/ZnVbcEL+UO/nUJxfNwwppS4y4kD/jb+/7
YZV2MKlIN6NdsRujIFdGkdBlNjafY7Xpm9STCMI3TLdwpstSaOvBh4+WmktalI0d
ey/ROvBCn/0=
=5L8+
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Clarence Gardner <clarence@king.silcom.com>
Subject: flexfax: Submitting job to hfaxd
To: flexfax@sgi.com
Date: Tue, 30 Sep 1997 19:37:51 -3100 (PDT)
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----


Hello, fax users!  I am trying to submit jobs to an hfaxd process.  All
I can find to go on is the list of commands in the man page and in the
Parser.c++ file.  So far, as you can see below, I'm using
	jnew, notifyaddr, dialstring, and dataformat, stot, and jsubm
and am getting the rather uninformative message "server nak'd request".
There are not errors logged on the server.  Is there a statement anywhere
about what is actually required to get a job submitted?  I've looked
in the archives without success.

Thanks for any help.

$ ./faxtest
*cmd* 'USER clarence'
*resp* '230 User clarence logged in.'
*cmd* 'jnew'
*resp* '200 New job created: jobid: 3062 groupid: 3062.'
*cmd* 'jparm notifyaddr clarence@silcom.com'
*resp* '213 NOTIFYADDR set to "clarence@silcom.com".'
*cmd* 'jparm dialstring 6859685'
*resp* '213 DIALSTRING set to "6859685".'
*cmd* 'jparm dataformat g32d'
*resp* '213 DATAFORMAT set to "g32d".'
*cmd* 'TYPE I'
*resp* '200 Type set to Image.'
*cmd* 'PORT 207,71,234,20,5,188'
*resp* '200 PORT command successful.'
*cmd* 'stot'
*resp* '150 FILE: /tmp/doc1517.ps (Opening new data connection).'
*resp* '226 Transfer complete (FILE: /tmp/doc1517.ps).'
*cmd* 'jsubm'
*resp* "460 Failed to submit job 3062: Unspecified reason
	 (scheduler NAK'd request)."


- -=-=-=-=-=-=-=-=
Clarence Gardner
AvTel Communications
Software Products and Services Division
clarence@silcom.com

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNDG3fAvaWr4ZT3PRAQE1bwQA06hMa8UP1Bk7QVF8Z1pDdNv+yMbb1ov/
GJi7cDoH8CsN+rVYBWUGmxRPrhXNrwdVbS8Dt2gV8GiHBu3FIgInyujMjgEbTXS7
U5FoaIo3fBhycbC+RbQ4Tj3qUhpNdUQ8mj0tKfMioRlcbn78y68fj5tABO72ofyx
2wGO/UykLXM=
=jCy3
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: "Guascor I+D" <guascor_id@jet.es>
To: <flexfax@sgi.com>
Cc: <kostka@sunbilbao.com>
Subject: flexfax: How to force pulse dialing (ATDP)
Date: Fri, 3 Oct 1997 12:05:51 +0100
X-MIME-Autoconverted: from 8bit to quoted-printable by minerva.ibernet.es id MAA00683
Sender: owner-flexfax@celestial.com

Hi,
I have installed hylafax latest version on two Sun sparc 5 running solaris
2.5.1. I got 2 Us Robotics Sporters voice 33.6 Faxmodem and Ive read the
HylaFax FAQ and trobleshooting many times. These machines are located in
two diferent locations, the only diference between them is that one uses
tone dialing while the other uses pulse dialing. My problem is that i don't
know how to tell hylafax to dial usin ATDP instead of ATDT. )
I've changed every single file containing the text ATDT (well, actually i
only found those files under etc and config ), but when i send a fax it
says NO CARRIER DETECTED and looking at the log file i see it stills dials
using ATDT.

I would appreciate very much any help

thanks in advance

Kostka Otaduy

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 03 Oct 1997 16:15:49 -0400
From: "Michael A. Irons" <bb851@lafn.org>
Organization: Yet another Linux machine via LA-FreeNet
To: flexfax@celestial.com
Subject: flexfax: HylaFAX-4.0pl1 (linux RPM) won't answer phone.
Sender: owner-flexfax@celestial.com

Hi all;

  After a disk crash I 'upgraded' to Red Hat 4.1 and installed the RPM
of HylaFAX-4.0pl1 from the binaries archive at SGI. I got it installed,
the servers come up fine. I can telent & ftp into them. Sendfax,queues
up fine. However I can't get faxgetty to answer the phone, even though 
the system getty and PPP work fine.
   Other then a new disk, HylaFAX worked fine with my slackware install
I had been using, so I know the modem works  with HylaFAX. After I use
faxconfig to set RingsBeforeAnswer to 2, I use kermit to check the modem
and S0=0 after every time. 
   I've looked in the mailing-list archive and nothing popped up. Could
someone please help me figure this out??

					Thanks
					Mike
-- 

"By the time they had diminished from 50 to 8, the other dwarves began
to suspect 'Hungry' ..."
                -- Gary Larson, "The Far Side"

		aTdHvAaNnKcSe	>THANKS 'in' advance<

					Michael A. Irons
					mailto:bb851@lafn.org

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Lou Katz <lou@metron.com>
Subject: flexfax: modem appears to be wedged
To: flexfax@sgi.com
Date: Fri, 3 Oct 1997 17:20:19 -0700 (PDT)
Sender: owner-flexfax@celestial.com

I just upgraded from Hylafax-3xx to 4.0p11. Running BSD/OS 2.1.
Modem is an old att-dataport 14.4.

Before the upgrade, I had one line (tty00) configured as bidir with rotary.

Incoming calls smoothly switched between data and fax. Outgoing faxes
worked fine. This line not really used for outging data.


After upgrade:

1. I get lots of "modem wedged" messages.
2. Outgoing faxes work fine
3. Outgoing pages work fine
4. Incoming faxes sort of work.
5. Incoming data never seems to connect.
6. After bad incoming try, modems doesn't answer the incoming ring for
   several minutes.

Any clues or pointers? I did RTFM.

Thanks
Lou Katz

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 06 Oct 1997 16:39:01 -0400 (EDT)
Organization: The Creative Group
From: Patrick Gardella <patrick@cre8tivegroup.com>
To: flexfax@celestial.com
Subject: flexfax: hfaxd quits
Sender: owner-flexfax@celestial.com

I'm running HylaFAX 4.0pl1 on FreeBSD 2.2.1 on P-200 with 128Megs RAM using
7 Multitech Multimodems 2834ZDX.

My hfaxd seemingly randomly quits.  There is nothing noted in syslog or anywhere
that I can find.  No cores anywhere. Where might I find the reason why it quits?

It's happened three times in the past week.

I start it from rc.d/hylafax.sh with:
/usr/local/sbin/hfaxd -i 4559 &

faxq and faxgetty keep going though.

Drives me crazy since I don't know that it quits until I start seeing error 
messages that faxes aren't going through.

Thanks!

Patrick

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 15 Oct 1997 15:47:19 --100
From: coppelli@sun3.dsea.unipi.it (Alessandro Coppelli)
To: flexfax@sgi.com
Subject: flexfax: Problem : any idea ?
Sender: owner-flexfax@celestial.com


 Hi to all.
 
  SCENARIO
  ========
  
   Linux box with USR Sportster modem/fax and latest Hylafax 
   
 PROBLEM
 =======
 
 
 
>Your facsimile job to 0,,055,367583 failed because:
>    Unspecified Transmit Phase C error; too many attempts to send
>
>    ---- Transcript of session follows ----
>
>May 21 13:12:41.90: [25709]: SESSION BEGIN 00000021 39055367583
>May 21 13:12:41.92: [25709]: SEND FAX: JOB 75 DEST 0,,055,367583 COMMID 
00000021
>May 21 13:12:41.92: [25709]: DELAY 2600 ms
>May 21 13:12:44.53: [25709]: <-- [15:ATE0V1Q0S0=0H0\r]
>May 21 13:12:44.66: [25709]: --> [2:OK]
>May 21 13:12:44.66: [25709]: <-- [21:ATS8=2S7=60&K3&D2&C1\r]
>May 21 13:12:44.80: [25709]: --> [2:OK]
>May 21 13:12:44.80: [25709]: <-- [12:AT+FCLASS=1\r]
>May 21 13:12:44.94: [25709]: --> [2:OK]
>May 21 13:12:44.94: [25709]: <-- [5:ATM0\r]
>May 21 13:12:45.08: [25709]: --> [2:OK]
>May 21 13:12:45.09: [25709]: <-- [12:AT+FCLASS=1\r]
>May 21 13:12:45.32: [25709]: --> [2:OK]
>May 21 13:12:45.33: [25709]: DIAL 0,,055,367583
>May 21 13:12:45.33: [25709]: <-- [19:ATDT0,,055,367583@\r]
>May 21 13:13:10.21: [25709]: --> [7:CONNECT]
>May 21 13:13:10.52: [25709]: --> [2:OK]
>May 21 13:13:10.52: [25709]: <-- [9:AT+FRH=3\r]
>May 21 13:13:10.54: [25709]: --> [7:CONNECT]
>May 21 13:13:10.83: [25709]: --> [2:OK]
>May 21 13:13:10.83: [25709]: REMOTE CSI "055367583"
>May 21 13:13:10.83: [25709]: <-- [9:AT+FRH=3\r]
>May 21 13:13:10.86: [25709]: --> [7:CONNECT]
>May 21 13:13:11.10: [25709]: --> [2:OK]
>May 21 13:13:11.10: [25709]: REMOTE best rate 9600 bit/s
>May 21 13:13:11.10: [25709]: REMOTE max page width 1728 pixels in 215 mm
>May 21 13:13:11.11: [25709]: REMOTE max unlimited page length 
>May 21 13:13:11.11: [25709]: REMOTE best vres 7.7 line/mm
>May 21 13:13:11.11: [25709]: REMOTE best format 1-D MR
>May 21 13:13:11.11: [25709]: REMOTE best 20 ms, 10 ms/scanline
>May 21 13:13:11.11: [25709]: USE 9600 bit/s
>May 21 13:13:11.11: [25709]: USE 20 ms, 10 ms/scanline
>May 21 13:13:11.11: [25709]: SEND file "docq/doc58.ps;01"
>May 21 13:13:11.11: [25709]: USE page width 1728 pixels in 215 mm
>May 21 13:13:11.11: [25709]: USE unlimited page length 
>May 21 13:13:11.11: [25709]: USE 7.7 line/mm
>May 21 13:13:11.11: [25709]: USE 1-D MR
>May 21 13:13:11.11: [25709]: SEND training at v.29 9600 bit/s
>May 21 13:13:11.11: [25709]: <-- [9:AT+FTH=3\r]
>May 21 13:13:11.15: [25709]: --> [7:CONNECT]
>May 21 13:13:12.65: [25709]: --> [7:CONNECT]
>May 21 13:13:13.15: [25709]: --> [2:OK]
>May 21 13:13:13.15: [25709]: DELAY 75 ms
>May 21 13:13:13.24: [25709]: <-- [10:AT+FTM=96\r]
>May 21 13:13:13.57: [25709]: --> [7:CONNECT]
>May 21 13:13:15.20: [25709]: --> [2:OK]
>May 21 13:13:15.20: [25709]: <-- [9:AT+FRH=3\r]
>May 21 13:13:16.45: [25709]: --> [7:CONNECT]
>May 21 13:13:16.60: [25709]: --> [2:OK]
>May 21 13:13:16.60: [25709]: TRAINING succeeded
>May 21 13:13:16.60: [25709]: <-- [10:AT+FTM=96\r]
>May 21 13:13:16.94: [25709]: --> [7:CONNECT]
>May 21 13:13:16.94: [25709]: SEND begin page
>May 21 13:13:31.03: [25709]: SENT 27495 bytes of data
>May 21 13:13:45.07: [25709]: SENT 27486 bytes of data
>May 21 13:13:49.40: [25709]: SENT 9463 bytes of data
>May 21 13:13:49.40: [25709]: SEND 1D RTC
>May 21 13:13:49.41: [25709]: SEND end page
>May 21 13:15:09.40: [25709]: <-- [9:AT+FTH=3\r]
>May 21 13:15:11.95: [25709]: --> [0:]
>May 21 13:15:11.96: [25709]: SEND FAILED: Unspecified Transmit Phase C error; 
too many attempts to send
>May 21 13:15:11.96: [25709]: <-- [5:ATH0\r]
>May 21 13:15:16.95: [25709]: MODEM <Timeout>
>May 21 13:15:16.97: [25709]: SESSION END
>
>    ---- Unsent job status ----


 What happaned ? Any idea ?
 
 Alessandro

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 15 Oct 1997 20:24:46 +0000
From: Didier Gautheron <dgautheron@magic.fr>
To: flexfax@sgi.com
Subject: Re: flexfax: Problem : any idea ?
Sender: owner-flexfax@celestial.com

Alessandro Coppelli wrote:
> 
>  Hi to all.
> 
>   SCENARIO
>   ========
> 
>    Linux box with USR Sportster modem/fax and latest Hylafax
> 
>  PROBLEM
>  =======
> 
> 
> 
> >Your facsimile job to 0,,055,367583 failed because:
> >    Unspecified Transmit Phase C error; too many attempts to send
> >
> >    ---- Transcript of session follows ----
> >
> >May 21 13:12:41.90: [25709]: SESSION BEGIN 00000021 39055367583
> >May 21 13:12:41.92: [25709]: SEND FAX: JOB 75 DEST 0,,055,367583 COMMID

> >May 21 13:12:44.94: [25709]: <-- [5:ATM0\r]

> >May 21 13:13:16.60: [25709]: --> [2:OK]
> >May 21 13:13:16.60: [25709]: TRAINING succeeded

> >May 21 13:13:49.40: [25709]: SEND 1D RTC
> >May 21 13:13:49.41: [25709]: SEND end page
> >May 21 13:15:09.40: [25709]: <-- [9:AT+FTH=3\r]
> >May 21 13:15:11.95: [25709]: --> [0:]
> >May 21 13:15:11.96: [25709]: SEND FAILED: Unspecified Transmit Phase C error;
> too many attempts to send
> >May 21 13:15:11.96: [25709]: <-- [5:ATH0\r]
> >May 21 13:15:16.95: [25709]: MODEM <Timeout>
> >May 21 13:15:16.97: [25709]: SESSION END
> >
> >    ---- Unsent job status ----
> 
>  What happaned ? Any idea ?
> 
>  Alessandro
Yea the other fax says you good bye (actually it hangs) after training
succeeded and before  SEND end of page. Remember faxes are half duplex
beasts. You could use ATM2 for sound and you'll hear it, when remote 'll
hang, sound will be  louder and you'll get the busy tone, with a little
practice it's easy - don't worry you'll qet the practice quickly :( -.
Ok.
Did you always have this problem with the same remote?
We had similar trouble in the past.
We had a little pause before sending the page (the idea's not from me)
and problems (at least this one) get away.

Didier.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: "Kevin Lai" <kevlaikf@mbox2.singnet.com.sg>
To: <flexfax@sgi.com>
Subject: flexfax: How to shorten delay when retry occurs?
Date: Tue, 14 Oct 1997 15:17:36 +0800
Sender: owner-flexfax@celestial.com

Whenever there is an error encountered by the fax server such as NO CARRIER
DETECTED, it will SLEEP for approximately 15 minutes.

I am using the compiled binaries for HP-UX 10.20, and the source codes does
not compile successfully.

Is the above configuration hardcoded to the binary files, or can it be
changed?
How can I configure it to use a shorter delay for retry?

Regards,

Kevin Lai

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: How to shorten delay when retry occurs?
To: kevlaikf@mbox2.singnet.com.sg (Kevin Lai)
Date: Thu, 16 Oct 1997 07:42:59 +0100 (BST)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> 
> Whenever there is an error encountered by the fax server such as NO CARRIER
> DETECTED, it will SLEEP for approximately 15 minutes.

If you are talking about a per job delay, rather than a per modem delay,
the default would appear to be more like 5 minutes (taken from config.h.in).
>From an old copy of the source code, this seems to be an undocumented 
parameter called jobreqother.

However, before changing it you should check the legal position in your
country.  Modems in the UK used to force a minimum interval between retries
of a number.  They no longer do, however the manuals require that modem
and software comply with a document, which I believe imposes the same
requirements.  The last set of these I saw allowed indefinite retries
at 10 minute intervals, but only allowed one fast retry and then only a
very limited number of tries in quite a few hours if the initial attempts
were at less than ten minute intervals.

The reasons for this are:

- silent or calling tone only wrong numbers can be very frightening;

- fast retries can cause denial of service;

- the telephone network is dimensioned for a particular number of 
  simultaneous calls and a particular attempt rate - to do anything else
  would put up the cost of calls, to pay for the extra equipment.

As an extreme example, my ISP used to provide service in Kingston upon
Hull, through a telephone switch owned by a third party.  Hull was the
one exception to the BT monopoly and BT is still not present there.
They had flat rate charging (charge per call) for local calls, and some
users abused this by staying connected all the time.  This meant there
were no free lines.

Eventually the company providing the lines cancelled their contract.
As far as I can tell this wasn't because of simultaneous call capacity
(erlangs), but because people repeat dialing to try to get in were
overloading their system.  (Note that a failed call may still reserve a
partial path through the network, as well as consuming processing resources,
tone decoders, etc.)

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 14 Oct 1997 20:40:35 -0500
From: Lachlan Dunlop <lach@lach.net>
To: flexfax@sgi.com
Subject: flexfax: trouble sending faxes??
Sender: owner-flexfax@celestial.com

Hello Again,

I have got my system up and running under Linux RH4.2, Multitech modem
with latest firmware 1432BA model.  Hylafax 4.0pl1.

I can receive faxes fine!!!!!  But when it comes to sending I'm all
messed up.

This is the /var/log/messages file:

.Oct 14 20:30:13 flipper FaxSend[18511]: MODEM MULTI-TECH SYSTEMS
MT1432BA/0116
Oct 14 20:30:13 flipper FaxSend[18511]: SEND FAX: JOB 7 DEST 4551326
COMMID 00000017
Oct 14 20:30:33 flipper FaxSend[18511]: Modem does not support document
page width, max page width 2432 pixels, image width 1734 pixels
Oct 14 20:30:43 flipper FaxSend[18520]: MODEM MULTI-TECH SYSTEMS
MT1432BA/0116
Oct 14 20:30:43 flipper FaxSend[18520]: SEND FAX: JOB 7 DEST 4551326
COMMID 00000018
Oct 14 20:31:03 flipper FaxSend[18520]: Modem does not support document
page width, max page width 2432 pixels, image width 1734 pixels
Oct 14 20:31:13 flipper FaxSend[18531]: MODEM MULTI-TECH SYSTEMS
MT1432BA/0116
Oct 14 20:31:13 flipper FaxSend[18531]: SEND FAX: JOB 7 DEST 4551326
COMMID 00000019
Oct 14 20:31:33 flipper FaxSend[18531]: Modem does not support document
page width, max page width 2432 pixels, image width 1734 pixels
Oct 14 20:31:35 flipper FaxQueuer[399]: JOB 7: REJECT: Too many attempts
to transmit: 3, max 3
Oct 14 20:31:35 flipper FaxQueuer[399]: NOTIFY: bin/notify "doneq/q7"
"rejected" ""
Oct 14 20:31:44 flipper FaxGetty[419]: MODEM MULTI-TECH SYSTEMS
MT1432BA/0116


Any Ideas would be appreciated.

Lach

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: "Gary Ng (Winclient)" <garyn@winclient.com>
To: flexfax@sgi.com, "'Lachlan Dunlop'" <lach@lach.net>
Subject: RE: flexfax: trouble sending faxes??
Date: Thu, 16 Oct 1997 17:11:04 +0800
Sender: owner-flexfax@celestial.com

	> page width, max page width 2432 pixels, image width 1734
pixels

^^^^^^^
	I thought there is only a number of valid width. For A4, it must
be 1728 ?

	Gary

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Comments: Authenticated sender is <michaelw@mail>
From: Michael Woodward <michaelw@solace.com.au>
Organization: Solace Technologies P/L
To: flexfax@sgi.com
Date: Thu, 16 Oct 1997 15:43:41 +1000
Subject: flexfax: Modem not answering when faxgetty running
Sender: owner-flexfax@celestial.com

I've just set up v4.0p11 on a SCO 5.0.0 system.  I have one modem 
connected to use for outgoing(only) faxes and incoming/outgoing data 
calls.

Outgoing faxes work fine, but I cannot dial in (data only).

It seems that while faxgetty is running, autoanswer on the modem is off 
and  the call does not get answered.
If I use a normal getty,  the modem answers but faxes just queue 
up and don't get sent.

Messages in the mail list archive refer to using `faxq -m modemdevice`, 
but this doesn't seem to be supported in this release ?

Can anyone help ??  Is more info required ??

---------------------------------------------------------------
Michael Woodward
Unix Systems Engineer           Phone: 03 9291 7523
Solace Technologies Pty Ltd     Fax  : 03 9291 7507
Level 11, 636 St. Kilda Road    Email: michaelw@solace.com.au
Melbourne, Victoria, Aust. 3004

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 16 Oct 1997 06:47:29 -0600 (MDT)
From: Dan Rosenblatt <dir@obo.org>
To: Michael Woodward <michaelw@solace.com.au>
cc: flexfax@sgi.com
Subject: Re: flexfax: Modem not answering when faxgetty running
Sender: owner-flexfax@celestial.com

On Thu, 16 Oct 1997, Michael Woodward wrote:
> I've just set up v4.0p11 on a SCO 5.0.0 system.  I have one modem 
> connected to use for outgoing(only) faxes and incoming/outgoing data 
> calls.
> 
> Outgoing faxes work fine, but I cannot dial in (data only).

There is (at least on my setup) a conflict between getty's
use of locking and HylaFax's use.  Add:
    LockDataCalls:          no
to your config.ttyXX file.  That fixed it for me.

-- C is to scalpel as C++ is to swiss-army chainsaw --

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
X-Authentication-Warning: tui.pinnacle.co.nz: jonc owned process doing -bs
Date: Fri, 17 Oct 1997 10:57:38 +1300 (NZDT)
From: Jonathan Chen <jonc@pinnacle.co.nz>
To: Michael Woodward <michaelw@solace.com.au>
cc: flexfax@sgi.com
Subject: Re: flexfax: Modem not answering when faxgetty running
Sender: owner-flexfax@celestial.com

On Thu, 16 Oct 1997, Michael Woodward wrote:

> I've just set up v4.0p11 on a SCO 5.0.0 system.  I have one modem
> connected to use for outgoing(only) faxes and incoming/outgoing data
> calls.
>
> Outgoing faxes work fine, but I cannot dial in (data only).
>
> It seems that while faxgetty is running, autoanswer on the modem is off
> and  the call does not get answered.

faxgetty(1m) actually monitors the RING responses generated by the modem
on an incoming call before picking up the line; ie the modem's autoanswer
*SHOULD* be off.

To enable incoming calls, check out `config.tty1A' for RingsBeforeAnswer:
and set this to a number > 0. Also make sure adaptive answer has been
enabled by use of the ModemSetupAACmd: parameter.

Additionally, for SCO 5.0.0, you should have

    GettyArgs: "-h -t60 %l o",

assuming you're running

    ModemRate: 38400

Cheers.
--
Jonathan Chen                          e-mail : jonc@pinnacle.co.nz
  Pinnacle Software Ltd                Voice  : +64.9.415.4460
  Auckland, New Zealand                Fax    : +64.9.415.4250

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 16 Oct 1997 18:00:39 +0800 (HKT)
From: Super-User <root@peach.asiainter.net>
Reply-To: Super-User <root@peach.asiainter.net>
To: flexfax@sgi.com
Subject: flexfax: postscript attachments show up at 5% size
Sender: owner-flexfax@celestial.com


hi, i'm playing with attaching postscript files and sending them through
hylafax.  when i do this, the header and body of the email come through
just fine, then i see some tiny thumbnail of the image that is supposed to
be the postscript page, in the bottom left corner of the first page.
i've tried saving the mail (with ps attachment) and running it through
faxmail (faxmail < ~/mail/sent-mail > output.ps) then using ghostscript or
solaris's pageview to view the output.  the postscript file i attached
appears in the bottom left corner, very very small (words become 1 or 2
pixels high).
i've played around with editing output.ps, and i see a line
".05 dup scale".  changing the .05 to .9 or 1 makes the attached
postscript file full page, and bumps the email header off.
changing it to 0.52 shows that it moves the header up, and gets cut in
half.

how are postscript attachments normally done?  is my postscript file
formatted to begin at the top of the page or what?  how can i have a
postscript attachment immediately following the header on the page?

also on examining the output i see two "showpage" and "end restore", this
results in an extra empty page.  (this comes up when there's an
attachment)..  any easy way to disable this?

thanks

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 16 Oct 1997 13:33:33 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Super-User <root@peach.asiainter.net>
Cc: flexfax@sgi.com
Subject: Re: flexfax: postscript attachments show up at 5% size
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Thu, 16 Oct 1997, Super-User wrote:

> hi, i'm playing with attaching postscript files and sending them through
> hylafax.  when i do this, the header and body of the email come through
> just fine, then i see some tiny thumbnail of the image that is supposed to
> be the postscript page, in the bottom left corner of the first page.
> i've tried saving the mail (with ps attachment) and running it through
> faxmail (faxmail < ~/mail/sent-mail > output.ps) then using ghostscript or
> solaris's pageview to view the output.  the postscript file i attached
> appears in the bottom left corner, very very small (words become 1 or 2
> pixels high).

Try using ps2fax, and look at the tiffg3 file generated with viewfax
on UNIX systems (from ftp://ftp.sgi.com/sgi/fax/contrib/).

> i've played around with editing output.ps, and i see a line
> ".05 dup scale".  changing the .05 to .9 or 1 makes the attached
> postscript file full page, and bumps the email header off.
> changing it to 0.52 shows that it moves the header up, and gets cut in
> half.

I suspect that your Postscript files are not compatible with the
ghostscript conversion. Take a look at the ps2fax script, and take
a look at your version of ghostscript or dps used to convert the
document.

> also on examining the output i see two "showpage" and "end restore", this
> results in an extra empty page.  (this comes up when there's an
> attachment)..  any easy way to disable this?

How are you generating the Postscript?

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNEZP2z/+ItycgIJRAQFoWgQAjcOTHhgB3gEAGFuXKg4XvYXwWOjJAuO4
pvzVcgr35ZqVgORrHwFexznJ1U3R2uu58Nj7WFCC5XJw7qWCdl7BQSZNnbPrtop5
tahi1ZbyvJc64J16rUIViKGrwB00NuWW/8k7kwMXfKws/8x1P6POjMMxbvMeLU3F
sXDYpkAFKOE=
=iMLO
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 29 Oct 1997 19:15:30 +0800 (HKT)
From: Super-User <root@peach.asiainter.net>
To: Ben Parker <ben@dha.unon.org>
cc: flexfax@sgi.com
Subject: Re: flexfax: postscript attachments show up at 5% size
Sender: owner-flexfax@celestial.com


On Wed, 29 Oct 1997, Ben Parker wrote:
> This a familiar problem. I have battled with faxmail for some time so that
> my e-mail-> fax gateway can handle attachments and I can tell you:
> 
> It's possible to send attachments through faxmail but you have to put in a
> couple of scripts for each mime type (ie image/gif). One to convert (ie
> GIF to PS), and one to massage the resulting output to compensate for
> faxmail trying to squash the image in 1/20 th original size in the bottom
> left corner.  Sending PS files is ironically the hardest battle. I have
tell me about it.. i've given up with this.

> discovered a silly workaroud. If you attach the PS file twice you get a
> thumbnail on Page 1 and a full size version on Page 2. Pretty punishing on
> transmit time, though.
well what i've been trying to do is send out email to fax in chinese
characters.  what i finally had to do is hack the hell out of faxmail so
it outputs PURE TEXT (get this), this is needed to properly decode
quoted-printable or whatever it's called, and decode the body if it's some
other encoding.  then run it through a program designed to convert chinese
text into bitmapped postscript.  then the result of that is fed into
sendfax.

your workaround doesn't work for me, coz the text/plain of every message
will be converted to ps.

...still looking for a prettier way...

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 29 Oct 1997 12:56:35 +0000
To: HylaFAX List <flexfax@sgi.com>
From: Phil Watkinson <pkw@elgro.co.uk>
Subject: Re: flexfax: postscript attachments show up at 5% size
Cc: Ben Parker <ben@dha.unon.org>
Sender: owner-flexfax@celestial.com

At 11:04 29/10/1997 -0300, Ben Parker wrote:
>[snip]
>            Sending PS files is ironically the hardest battle. I have
>discovered a silly workaroud. If you attach the PS file twice you get a
>thumbnail on Page 1 and a full size version on Page 2.
>[snip]
>I guess the following from faxmail(1) might be the clue to sorting this
>out and starting each part of a multipart message on a new page. Can
>anyone PLEASE suggest what Postscript command would work here (would
>"showpage"  do it?)
>
>DigestDivider     string       -                    multipart/digest
divider POSTSCRIPT command
>

The other suspect is the faxmail prologue file, faxmail.ps. It says :-

      %       $Id: faxmail.c++,v 1.56 1996/07/02 19:49:58 sam Exp sam
      %
      % faxmail prologue.
      %
      /EPSbegin{save gsave 20 20 scale 200 dict begin}def
      /EPSend{end grestore restore}def

Note the '20 20 scale' in the macro named 'EPSbegin' which scales
the image 20 times in both the x and y axis.

I would hazard a guess that the two macroes are meant to be wrapped
around an EPS file when it is inserted into postscript. Whether that
means that any attachments should be so suffixed and prefixed; or
whether it is a mechanism for something else is an important question.

I had not found any occasion of either EPSbegin or EPSend being called
in any postscript generated by faxmail for email without attachments and 
for email with an single attachment.  Your comment about a second 
attachment being at the correct scale begs the question whether these 
macroes have been generated in the postscript of the second copy?

If they have - fingers crossed - the problem then becomes to find out why 
they weren't called for the first copy of the attachment.

As you can probably tell, I'm guessing. Apologies if it hinders rather
than helps.

Cheers,
Phil Watkinson,
Boston, UK.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 29 Oct 1997 11:04:51 -0300 (GMT+3)
From: Ben Parker <ben@dha.unon.org>
Reply-To: Ben Parker <ben@dha.unon.org>
To: Super-User <root@peach.asiainter.net>
cc: flexfax@sgi.com
Subject: Re: flexfax: postscript attachments show up at 5% size
Sender: owner-flexfax@celestial.com

This a familiar problem. I have battled with faxmail for some time so that
my e-mail-> fax gateway can handle attachments and I can tell you:

It's possible to send attachments through faxmail but you have to put in a
couple of scripts for each mime type (ie image/gif). One to convert (ie
GIF to PS), and one to massage the resulting output to compensate for
faxmail trying to squash the image in 1/20 th original size in the bottom
left corner.  Sending PS files is ironically the hardest battle. I have
discovered a silly workaroud. If you attach the PS file twice you get a
thumbnail on Page 1 and a full size version on Page 2. Pretty punishing on
transmit time, though.

I guess the following from faxmail(1) might be the clue to sorting this
out and starting each part of a multipart message on a new page. Can
anyone PLEASE suggest what Postscript command would work here (would
"showpage"  do it?)

DigestDivider     string       -                    multipart/digest divider POSTSCRIPT command

Cheers,

Ben
Nairobi
 


On Thu, 16 Oct 1997, Super-User wrote:

> 
> hi, i'm playing with attaching postscript files and sending them through
> hylafax.  when i do this, the header and body of the email come through
> just fine, then i see some tiny thumbnail of the image that is supposed to
> be the postscript page, in the bottom left corner of the first page.
> i've tried saving the mail (with ps attachment) and running it through
> faxmail (faxmail < ~/mail/sent-mail > output.ps) then using ghostscript or
> solaris's pageview to view the output.  the postscript file i attached
> appears in the bottom left corner, very very small (words become 1 or 2
> pixels high).
> i've played around with editing output.ps, and i see a line
> ".05 dup scale".  changing the .05 to .9 or 1 makes the attached
> postscript file full page, and bumps the email header off.
> changing it to 0.52 shows that it moves the header up, and gets cut in
> half.
> 
> how are postscript attachments normally done?  is my postscript file
> formatted to begin at the top of the page or what?  how can i have a
> postscript attachment immediately following the header on the page?
> 
> also on examining the output i see two "showpage" and "end restore", this
> results in an extra empty page.  (this comes up when there's an
> attachment)..  any easy way to disable this?
> 
> thanks
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Louis REINARD <lrei@ictlux.com>
Date: Thu, 30 Oct 1997 14:13:10 +0100
To: flexfax@sgi.com
Cc: crei@ictlux.com
Subject: Re: flexfax: postscript attachments show up at 5% size
X-Parental-advisory: Explicit lyrics
X-Nic-hdl: LR9
X-Phone: +352-295383-242
X-Fax: +352-295383-222
X-Disclaimer: Any opinions in this message are my own and not those of my present or previous employer(s).
X-No-Junk-Mail: I do not want to get *any* junk mail.
Sender: owner-flexfax@celestial.com

Hello,

On Wed, Oct 29, 1997 at 11:04:51AM -0300, Ben Parker wrote:
> It's possible to send attachments through faxmail but you have to put in a
> couple of scripts for each mime type (ie image/gif). One to convert (ie
> GIF to PS), and one to massage the resulting output to compensate for
> faxmail trying to squash the image in 1/20 th original size in the bottom
> left corner.  Sending PS files is ironically the hardest battle. I have
> discovered a silly workaroud. If you attach the PS file twice you get a
> thumbnail on Page 1 and a full size version on Page 2. Pretty punishing on
> transmit time, though.

I'm running hylafax-v4.0pl1 on a Linux 2.0.29 system.
I have that atatchment problem too, and after some investigation,
I found the following:

For an email-message like this:

>  To: x999
>  From <lrei@ictlux.com>
>  Subject: test

  The quick brown fox...

faxmail produces a PS output like this:

  %!PS-Adobe-3.0
  %%Creator: HylaFAX TextFmt Class
  %%Title: HylaFAX-Mail
[...snip...]
  %       $Id: faxmail.c++,v 1.56 1996/07/02 19:49:58 sam Exp sam
  %
  % faxmail prologue.
  %
  /EPSbegin{save gsave 20 20 scale 200 dict begin}def
  /EPSend{end grestore restore}def
  end
  %%EndProlog
  %%Page: "1" 1
  save $printdict begin
  .05 dup scale
   sf0 ()[()()](1)Header
  360 16092 moveto
  B(To:)s1  570 M (x999)s2 EL
  B(Subject:)s1  94 M (test)s2 EL
  B EL
  B(The quick brown fox...)S

  showpage
  end restore
  %%Trailer
  %%Pages: 1
  %%EOF

Please pay attention to the `.05 dup scale' and the last 5 lines.

Now I take an email message like this:

>  To: x999
>  From <lrei@ictlux.com>
>  Subject: test
>  MIME-Version: 1.0
>  Content-Type: multipart/mixed; boundary="------------6285531E7FBD"
>
>  This is a multi-part message in MIME format.
>
>  --------------6285531E7FBD
>  Content-Type: text/plain; charset=us-ascii
>  Content-Transfer-Encoding: 7bit
>
>  http://www.w3.org/ 
>
>  --------------6285531E7FBD
>  Content-Type: text/html; charset=us-ascii
>  Content-Transfer-Encoding: 7bit
>  Content-Base: "http://www.w3.org/"
>
>  <BASE HREF="http://www.w3.org/">
>
>  <HTML><HEAD>
>[...snip...]
>  </HTML>
>
>  --------------6285531E7FBD--

and faxmail produces this PS:

  %!PS-Adobe-3.0
  %%Creator: HylaFAX TextFmt Class
  %%Title: HylaFAX-Mail
[...snip...]
  %       $Id: faxmail.c++,v 1.56 1996/07/02 19:49:58 sam Exp sam
  %
  % faxmail prologue.
  %
  /EPSbegin{save gsave 20 20 scale 200 dict begin}def
  /EPSend{end grestore restore}def
  end
  %%EndProlog
  %%Page: "1" 1
  save $printdict begin
  .05 dup scale
   sf0 ()[()()](1)Header
  360 16092 moveto
  B(To:)s1  570 M (x999)s2 EL
  B(From:)s1  316 M (<lrei@incotech.lu>)s2 EL
  B(Subject:)s1  94 M (test)s2 EL
  B EL
  B(http://www.w3.org/)S
  B()S
  %!
  %%Title: W3
  %%Creator: html2ps version 1.0 beta1
  %%CreationDate: Thu Oct 30 12:54:21 1997
[...snip...]
  end restore

  showpage
  end restore
  %%Trailer
  %%Pages: 1
  %%EOF

Again, pay attention to the `.05 dup scale', the last 5 lines AND the
position where the output of the text/html MIME-converter-script
(using html2ps in my case) has been inserted.

As far as I can see, the output from the converter-script has
not been appended to the PS generated from the first part of the
multipart MIME message, but inserted before the `showpage'
`end restore' of that first page. The result of this is, that
the `.05 dup scale' is still in effect when the converter output
gets interpreted. This results in the small "thumbnail" on the
first page that a few of you have described.

Also, this explains why the "double attachments" described by
Ben Parker work. The first attachment gets scaled down because
of the reason descibed above. This first attachment supposedly
ends with `showpage' and `end restore' which put the effect of
`.05 dup scale' to an end, and then his second attachment gets
faxed OK.

I have manually moved those last five lines so that they precede
the start `%!' of the converter-script output and then the fax
just looks fine.

As it appears to me, the problem is situated in faxmail or the
TextFmt routines used by faxmail. Due to my lack of c/c++
programming skills, I have not been able to hack the source
to get this solved.

Hope this helps.
Louis
-- 
   Louis REINARD - LX 1 QA
   At work: lrei@ictlux.com , PGP KeyID: 1024/0xEB16815D
   At home: lou@ictlux.com  , PGP KeyID: 1024/0x1F5F4EFD
   To get PGP public keys, email with Subject: SENDPGPKEY

   Misfortune: (n), a fortune that never misses.
                -- Ambrose Bierce, The devil's dictionary


From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 31 Oct 1997 13:03:43 +0800 (HKT)
From: Super-User <root@peach.asiainter.net>
To: Louis REINARD <lrei@ictlux.com>
cc: flexfax@sgi.com, crei@ictlux.com
Subject: Re: flexfax: postscript attachments show up at 5% size
Sender: owner-flexfax@celestial.com



yes, i figured it was something like that - although i would really like
the attachments to be shown in the same page as the to:/from: header (the
body of the message would be converted into an attachment)..

has anyone had trouble with faxmail core dumping on mail with many parts,
such as a forward from netscape communicator's mail, or a returned mail
(user/host unknown) ?  this happens all the time here.


On Thu, 30 Oct 1997, Louis REINARD wrote:
> As far as I can see, the output from the converter-script has
> not been appended to the PS generated from the first part of the
> multipart MIME message, but inserted before the `showpage'
> `end restore' of that first page. The result of this is, that
> the `.05 dup scale' is still in effect when the converter output
> gets interpreted. This results in the small "thumbnail" on the
> first page that a few of you have described.
> 
> Also, this explains why the "double attachments" described by
> Ben Parker work. The first attachment gets scaled down because
> of the reason descibed above. This first attachment supposedly
> ends with `showpage' and `end restore' which put the effect of
> `.05 dup scale' to an end, and then his second attachment gets
> faxed OK.
> 
> I have manually moved those last five lines so that they precede
> the start `%!' of the converter-script output and then the fax
> just looks fine.
> 
> As it appears to me, the problem is situated in faxmail or the
> TextFmt routines used by faxmail. Due to my lack of c/c++
> programming skills, I have not been able to hack the source
> to get this solved.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 31 Oct 1997 18:30:10 +0000
To: HylaFAX List <flexfax@sgi.com>
From: Phil Watkinson <pkw@elgro.co.uk>
Subject: Re: flexfax: postscript attachments show up at 5% size
Cc: Louis REINARD <lrei@ictlux.com>
Sender: owner-flexfax@celestial.com


Louis REINARD <lrei@ictlux.com> wrote:

>[snip some very clear and useful analytical work]
>Also, this explains why the "double attachments" described by
>Ben Parker work. The first attachment gets scaled down because
>of the reason descibed above. This first attachment supposedly
>ends with `showpage' and `end restore' which put the effect of
>`.05 dup scale' to an end, and then his second attachment gets
>faxed OK.
>
>I have manually moved those last five lines so that they precede
>the start `%!' of the converter-script output and then the fax
>just looks fine.
>
>As it appears to me, the problem is situated in faxmail or the
>TextFmt routines used by faxmail. Due to my lack of c/c++
>programming skills, I have not been able to hack the source
>to get this solved.
>

Good news. We seem to be on the trail of this common bug. My
thought that it was something to do with the faxmail prologue
file was obviously a complete miss.

Which is a bit of a shame, since I was hoping that if the solution
lay within the prologue file, then it would be easy to enhance
faxmail to provide an user defined header on the fax.

Can I give you an example ?

With a prologue file such as the following ( borrowed freely from
faxcover.ps ) :

%       $Id: faxmail.c++,v 1.56 1996/07/02 19:49:58 sam Exp sam
%
% faxmail prologue.
%
/inch {72 mul} def
/userheader {
gsave
  4.4 inch 10.2 inch moveto
  /Helvetica-BoldOblique findfont 23 scalefont setfont
  (Elgro Limited) show
  4.42 inch 9.95 inch moveto
  /Helvetica-Bold findfont 14 scalefont setfont
  (Boston, UK) show
grestore
} def

And then fiddling about with the faxmail generated postscript file to 
call the 'userheader' macro - after where the prologue file is inserted,
but *before* the infamous '0.5 dup scale' line : 

[snip, snip, start of ps file ]
                        <-----this is where prologue file is inserted
%%EndProlog
%%Page: "1" 1
userheader              <-----this is where user defined macro is called
save $printdict begin
.05 dup scale
[snip, snip, rest of ps file]

Then the result is a rather nice heading on the top right of the fax. So,
while someone is hacking faxmail or textfmt, could this facility be added
as well??

In my enthusiasm, I started looking through TextFmt.c++, but all I could
see was data structures, definitions and whatnot  -  couldn't find any
code that actually *did* anything. Clever stuff this c++, and well
beyond me.

Sorry for the long post,

Phil Watkinson,
Boston, UK.

From lrei@ictlux.com  Wed Feb 17 13:55:47 1999
From: Louis REINARD <lrei@ictlux.com>
Date: Sat, 1 Nov 1997 13:44:36 +0100
To: Phil Watkinson <pkw@elgro.co.uk>
Cc: HylaFAX List <flexfax@sgi.com>
Subject: Re: flexfax: postscript attachments show up at 5% size
X-Parental-advisory: Explicit lyrics
X-Nic-hdl: LR9
X-Phone: +352-295383-242
X-Fax: +352-295383-222
X-Disclaimer: Any opinions in this message are my own and not those of my present or previous employer(s).
X-No-Junk-Mail: I do not want to get *any* junk mail.

On Fri, Oct 31, 1997 at 06:30:10PM +0000, Phil Watkinson wrote:
> Good news. We seem to be on the trail of this common bug.
I think so. Have a look at facmail.c++:

[...snip...]
    beginFile();

    formatHeaders(*this);               // format top-level headers

    const fxStr* version = findHeader("MIME-Version");
    if (version && *version == "1.0")
        formatMIME(stdin, mime, *this); // parse MIME format
    else
        formatText(stdin, mime);        // treat body as text/plain

    endFile();
    endFormatting();
[...snip...]

With the help of a C++ speaking colleague, I found that the "last 5 lines"

  showpage                                                                 
  end restore                                                            
  %%Trailer                                                              
  %%Pages: 1                                                                
  %%EOF

which I described in my last mail, get generated in the endFile() routine.
That is AFTER the inclusion of any PS that got
generated by an external converter-script which happens inside
formatMIME(). The code above is OK if the mail to be faxed is only text
or a MIME-message with a single text/plain part.

As soon as you have a MIME multi-part, f.ex. where the first part is
text/plain and the second part has to go through a converter-script,
let's say text/html, these "5 lines" should end the postscript 
generated from the first part, the text/plain in this example. Then
should follow the postscript for the text/html.

In other words, the postscript-code for the first page (the page with
To:, From:, etc. header lines) is not "finished" (the "5 lines" are
still missing) when the postscript code for other MIME parts gets
generated. Then these "5 lines" get added at the end of the postscript
code when all MIME-parts have been converted.

>From the facts at hand, I conclude that these "5 lines" get generated
at the wrong moment.

> And then fiddling about with the faxmail generated postscript file to 
> call the 'userheader' macro - after where the prologue file is inserted,
> but *before* the infamous '0.5 dup scale' line : 
> 
> [snip, snip, start of ps file ]
>                         <-----this is where prologue file is inserted
> %%EndProlog
> %%Page: "1" 1
> userheader              <-----this is where user defined macro is called
> save $printdict begin
> .05 dup scale
> [snip, snip, rest of ps file]
> 
> Then the result is a rather nice heading on the top right of the fax. So,
> while someone is hacking faxmail or textfmt, could this facility be added
> as well??
> 
> In my enthusiasm, I started looking through TextFmt.c++, but all I could
> see was data structures, definitions and whatnot  -  couldn't find any
> code that actually *did* anything. Clever stuff this c++, and well
> beyond me.

Good idea, I'll have to try that...
You were looking at the right place, I think.
This is from TextFmt.c++:

[...snip...]
void
TextFmt::beginCol(void)
{
    if (column == 1) {                          // new page
        if (reverse)  {
            int k = pageNum-firstPageNum;
            off_t off = (off_t) ftell(tf);
            if (k < pageOff->length())
                (*pageOff)[k] = off;
            else
                pageOff->append(off);
        }
        fprintf(tf,"%%%%Page: \"%d\" %d\n", pageNum-firstPageNum+1, pageNum);
        fputs("save $printdict begin\n", tf);
        fprintf(tf, ".05 dup scale\n");
        curFont->setfont(tf);
        if (landscape)
            fputs("LandScape\n", tf);
[...snip...]

It is the place where the "infamous" `.05 dup scale' gets generated.
There should be no problem to add a line to insert your `userheader'.
But there is a catch here: TextFmt::beginCol(void) is called for
the begin of a column, which is not identical with the begin of a
page. As you know, a few of the HylaFax utils, a.o. faxmail, can be
started with `-1' or `-2' in the command-line optins to set text in
one or two columns. As long a the default (1  column) is used, begin
of column and begin of page is "the same". I have not yet played
with text set multi-column. All my fiddling in the code was done
with text set single-column.

Hope this helps,
Have a nice weekend.

Louis
-- 
   Louis REINARD - LX 1 QA
   At work: lrei@ictlux.com , PGP KeyID: 1024/0xEB16815D
   At home: lou@ictlux.com  , PGP KeyID: 1024/0x1F5F4EFD
   To get PGP public keys, email with Subject: SENDPGPKEY

   Misfortune: (n), a fortune that never misses.
                -- Ambrose Bierce, The devil's dictionary

Attachment Converted: "C:\PROGRAM FILES\EUDORA\Attach\Re flexfax postscript attachm2"

From ben@dha.unon.org  Wed Feb 17 13:55:47 1999
Date: Mon, 3 Nov 1997 09:37:46 -0300 (GMT+3)
From: Ben Parker <ben@dha.unon.org>
To: Louis REINARD <lrei@ictlux.com>
cc: Phil Watkinson <pkw@elgro.co.uk>, HylaFAX List <flexfax@sgi.com>
Subject: Re: flexfax: postscript attachments show up at 5% size


Something like the script below could "eject" a page before faxing the
attachment.  This catches this text in the output of html2ps (but might
screw up witha document with the word "variable" in!) (By the way, my
mileage is varying wildly on html2ps. Is there any other tool out there?). 

Unfortunately there doesn't seem to be a common place in PS files that you
can insert this, so each converter script needs a companion "ejecter"
script tailored to the peculiarities of the PS out from whatever prog you
used.

html2ps conveniently has this line in the output which you can catch and
eject before...

%--End of variable part --

so the script can be something like:

#!/usr/bin/perl
while (<STDIN>) {
	if ($_ =~ "variable") {
		print "showpage\n";
	}
	print $_;  
} 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: edw@detel.com (Ed Weinberg)
To: flexfax@sgi.com
Subject: flexfax: RPM: document conversion script was not found
Date: Thu, 23 Oct 1997 03:35:42 GMT
Reply-To: edw@detel.com
Sender: owner-flexfax@celestial.com

Getting closer!   

There were too many things wrong with my install.  I don't know if it
was caused by upgrading rpm's or what, but I uninstalled then
re-installed the hylafax rpm and got MUCH farther...

Here is the current, and maybe last, problem.  

This is from the failure message emailed to me:
Your facsimile job to 912033674005 was not sent because
the document conversion script was not found.


This is from /var/log/messages:
Oct 22 21:53:16 detel FaxQueuer[407]: JOB 2: CONVERT DOCUMENT: exit
status 0xff00
Oct 22 21:53:17 detel FaxQueuer[226]: NOTIFY: bin/notify "doneq/q2"
"no_formatter" ""
Oct 22 21:55:44 detel FaxQueuer[559]: No regular expression for modem
class

Can anyone shed some light on this?  Do I have two problems?

  --  Ed Weinberg,
      Detel, Inc., An Internet Presence Provider
      edw@detel.com
      www.detel.com/
      www.serverking.com
      www.q5.com/  <-- find someone to CoolTalk or chat with here

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 11:30:09 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Ed Weinberg <edw@detel.com>
Cc: flexfax@sgi.com
Subject: Re: flexfax: RPM: document conversion script was not found
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Thu, 23 Oct 1997, Ed Weinberg wrote:

> There were too many things wrong with my install.  I don't know if it
> was caused by upgrading rpm's or what, but I uninstalled then
> re-installed the hylafax rpm and got MUCH farther...
> 
> Here is the current, and maybe last, problem.  
> 
> This is from the failure message emailed to me:
> Your facsimile job to 912033674005 was not sent because
> the document conversion script was not found.
> 
> 
> This is from /var/log/messages:
> Oct 22 21:53:16 detel FaxQueuer[407]: JOB 2: CONVERT DOCUMENT: exit
> status 0xff00
> Oct 22 21:53:17 detel FaxQueuer[226]: NOTIFY: bin/notify "doneq/q2"
> "no_formatter" ""
> Oct 22 21:55:44 detel FaxQueuer[559]: No regular expression for modem
> class
> 
> Can anyone shed some light on this?  Do I have two problems?

Hmm. Did you send a plain text or a Postscript document or a TIFF or a
tiffg3 document? I suspect the Ghostscript installation of having a
potential difficulty.

I also assume that "faxstat" now works correctly?


			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNE9tej/+ItycgIJRAQFUZQP+MJZ1z9b9EphDzgvHM8fDmRrf4x/lNWGi
Ty4Ea2VsQg4ImWcNjrHDMGu1XxJWOp9FemZDCvfD27qA2uuLqF+ZiLbTPy9IQ0Fh
+PYpwqUE71+AENDLjcLr84P4CflmxQ3ScZz8/pVPs4lp6QdjjWb+XVV3t+hVuC4a
Ypgb2ubYgwo=
=4MK6
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Steve Williams <steve@genie96.com>
Subject: Re: flexfax: RPM: document conversion script was not found
To: edw@detel.com
Date: Thu, 23 Oct 1997 09:41:52 -0600 (MDT)
Cc: flexfax@sgi.com
Read-Receipt-To: steve@genie96.com
Sender: owner-flexfax@celestial.com

[snip]

God, I love it!!.  

I used this EXACT problem as an example of the power of this mailing list
a couple of days ago re: this ongoing "Linux RPM issue", and the reasons
to not be editing the config files by hand INITIALLY.

Somewhere on your system there should be a plain old "config" file. 
( /usr/spool/fax/etc/config on my system ) Not the Modem config file, but one
of the the HylaFAX config files.

In there, there is a line "ModemClass:" line.  I will give you odds
that you put something like 1 or 2 or 2.0 in that line.

This is NOT what it is for.  Delete the line unless you RTFM what this
paramater is for.  If you are going to use it, it should look something
like:

ModemClass:             "any: tty[25]"

This is the danger of not accepting the defaults of "faxaddmodem", which
I think would have left this field blank.

Good Luck.
> 
> This is from /var/log/messages:
> Oct 22 21:53:16 detel FaxQueuer[407]: JOB 2: CONVERT DOCUMENT: exit
> status 0xff00
> Oct 22 21:53:17 detel FaxQueuer[226]: NOTIFY: bin/notify "doneq/q2"
> "no_formatter" ""
> Oct 22 21:55:44 detel FaxQueuer[559]: No regular expression for modem
> class
> 
> Can anyone shed some light on this?  Do I have two problems?
> 
>   --  Ed Weinberg,
>       Detel, Inc., An Internet Presence Provider
>       edw@detel.com
>       www.detel.com/
>       www.serverking.com
>       www.q5.com/  <-- find someone to CoolTalk or chat with here
> 
> 


-- 
	Steve Williams, Calgary, Alberta, Canada
	Genie Computer Systems Inc.
	steve@genie96.com

"A man doesn't begin to attain wisdom until he recognizes that he is 
 no longer indespensable."
- Admiral Richard E. Byrd ( 1888-1957 )

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: edw@detel.com (Ed Weinberg)
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
Cc: flexfax@sgi.com
Subject: Re: flexfax: RPM: document conversion script was not found
Date: Thu, 23 Oct 1997 16:07:28 GMT
Reply-To: edw@detel.com
Sender: owner-flexfax@celestial.com

On Thu, 23 Oct 1997 11:30:09 -0400 (EDT), Nico Garcia
<raoul@cirl.meei.harvard.edu> wrote:
>On Thu, 23 Oct 1997, Ed Weinberg wrote:
>
>> Here is the current, and maybe last, problem.  
>> 
>> This is from the failure message emailed to me:
>> Your facsimile job to 912033674005 was not sent because
>> the document conversion script was not found.
>
>Hmm. Did you send a plain text or a Postscript document or a TIFF or a
>tiffg3 document? I suspect the Ghostscript installation of having a
>potential difficulty.
Text file.  I used the command:
sendfax -d 912033674005 /etc/hosts

>I also assume that "faxstat" now works correctly?
Yes.


  --  Ed Weinberg,
      Detel, Inc., An Internet Presence Provider
      edw@detel.com
      www.detel.com/
      www.serverking.com
      www.q5.com/  <-- find someone to CoolTalk or chat with here

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: edw@detel.com (Ed Weinberg)
To: Steve Williams <steve@genie96.com>
Cc: flexfax@sgi.com
Subject: Re: flexfax: RPM: document conversion script was not found
Date: Thu, 23 Oct 1997 19:23:35 GMT
Reply-To: edw@detel.com
Sender: owner-flexfax@celestial.com

On Thu, 23 Oct 1997 09:41:52 -0600 (MDT), Steve Williams
<steve@genie96.com> wrote:

We get even closer!

>I used this EXACT problem as an example of the power of this mailing list
>a couple of days ago re: this ongoing "Linux RPM issue", and the reasons
>to not be editing the config files by hand INITIALLY.
>
>Somewhere on your system there should be a plain old "config" file. 
>( /usr/spool/fax/etc/config on my system ) Not the Modem config file, but one
>of the the HylaFAX config files.
/var/spool/hyla/etc/config

>In there, there is a line "ModemClass:" line.  I will give you odds
>that you put something like 1 or 2 or 2.0 in that line.
It came that way.  I just commented it out with a "#", rebooted (did I
need to reboot?), then used the command:
sendfax -d 912033674005 /etc/hosts
In the hope that it would fax the text document to 9+1+203-367-4005.
It bombed.
The email announceing the bomb status was the same as before.  There
may be one less error message in /var/log/messages.

>ModemClass:             "any: tty[25]"
Should I try this line?  ^^

Here is what config currently has in it.  I have not even changed what
I think is the area code.

LogFacility:            daemon
CountryCode:            1
AreaCode:               999
LongDistancePrefix:     1
InternationalPrefix:    011
DialStringRules:        "etc/dialrules"
ServerTracing:          1
#ModemClass:            2

  --  Ed Weinberg,
      Detel, Inc., An Internet Presence Provider
      edw@detel.com
      www.detel.com/
      www.serverking.com
      www.q5.com/  <-- find someone to CoolTalk or chat with here

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Steve Williams <steve@genie96.com>
Subject: Re: flexfax: RPM: document conversion script was not found
To: edw@detel.com
Date: Thu, 23 Oct 1997 13:56:26 -0600 (MDT)
Cc: flexfax@sgi.com
Read-Receipt-To: steve@genie96.com
Sender: owner-flexfax@celestial.com

> 
> On Thu, 23 Oct 1997 09:41:52 -0600 (MDT), Steve Williams
> <steve@genie96.com> wrote:
> 
> We get even closer!
> 
[snip]

> >that you put something like 1 or 2 or 2.0 in that line.
> It came that way.  I just commented it out with a "#", rebooted (did I
> need to reboot?), then used the command:
> sendfax -d 912033674005 /etc/hosts
> In the hope that it would fax the text document to 9+1+203-367-4005.
> It bombed.
> The email announceing the bomb status was the same as before.  There
> may be one less error message in /var/log/messages.
> 

That is VERY important that there is one less line in the messages file.
That means that deleteing the line fixed that problem.

I would need to see the logfile ( I delete all my mail after reading <g> )
but the problem is that in your $FAXSPOOL/bin, there should be a program
called ps2fax, that should fire up ghostscript.  Either it's missing,
or what it is doing is wrong.  Try to run this program by hand.  There is
a man page ( I believe ) for ps2fax ( NOTE I believe you have to be in the 
$FAXSPOOL directory when you run ps2fax )

All that said, 

I would be inclined to make a copy of your "config" & config.modem files
someplace safe, remove them, and run "faxsetup".  If you don't know how
to answer a question, just accept the default.

This will fully test out your system configuration and make sure things are
what they should be.

At the end, it will ask to add a modem, and I would say yes to that.  If
it works, then your system should be ready to go.  

In the end, this will save you MUCH agony, even if faxaddmodem won't run,
it is worthwhile running faxsetup.

-- 
	Steve Williams, Calgary, Alberta, Canada
	Genie Computer Systems Inc.
	steve@genie96.com

"A man doesn't begin to attain wisdom until he recognizes that he is 
 no longer indespensable."
- Admiral Richard E. Byrd ( 1888-1957 )

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 16:30:12 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Ed Weinberg <edw@detel.com>
Cc: flexfax@sgi.com
Subject: Re: flexfax: RPM: document conversion script was not found
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Thu, 23 Oct 1997, Ed Weinberg wrote:

> On Thu, 23 Oct 1997 11:30:09 -0400 (EDT), Nico Garcia
> <raoul@cirl.meei.harvard.edu> wrote:
> >On Thu, 23 Oct 1997, Ed Weinberg wrote:
> >
> >> Here is the current, and maybe last, problem.  
> >> 
> >> This is from the failure message emailed to me:
> >> Your facsimile job to 912033674005 was not sent because
> >> the document conversion script was not found.
> >
> >Hmm. Did you send a plain text or a Postscript document or a TIFF or a
> >tiffg3 document? I suspect the Ghostscript installation of having a
> >potential difficulty.
> Text file.  I used the command:
> sendfax -d 912033674005 /etc/hosts

OK. Pre-process it: use something like textfmt to turn the text into
Postscript by hand, then turn the Postscript into tiffg3 with ps2fax and
see where things break....

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNE+zzT/+ItycgIJRAQGH7wP/QtfLXWRU1kp6lz7q/JATaYfSEbav84l8
h3GKzD7nQbWbqcOw9aZjGllpXJbff3gIHgUfp2ezCDVySEdeCPj6ZfCLd8aokK5y
hTGrjf4xC//szR6SgFLkpJX7SA9dYoCCBr93o//WVLqEvrdJLVrwnekjKhwKJ4HY
br+w1BuSMfk=
=51el
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: edw@detel.com (Ed Weinberg)
To: Steve Williams <steve@genie96.com>
Cc: flexfax@sgi.com
Subject: Re: flexfax: RPM: document conversion script was not found
Date: Thu, 23 Oct 1997 21:03:11 GMT
Reply-To: edw@detel.com
Sender: owner-flexfax@celestial.com

On Thu, 23 Oct 1997 13:56:26 -0600 (MDT), Steve Williams
<steve@genie96.com> wrote:

>> >that you put something like 1 or 2 or 2.0 in that line.
>> It came that way.  I just commented it out with a "#", rebooted (did I
>> need to reboot?), then used the command:
>> sendfax -d 912033674005 /etc/hosts
>> In the hope that it would fax the text document to 9+1+203-367-4005.
>> It bombed.

>That is VERY important that there is one less line in the messages file.
>That means that deleteing the line fixed that problem.
I guessed that.

>I would need to see the logfile ( I delete all my mail after reading <g> )
>but the problem is that in your $FAXSPOOL/bin
Should I define $FAXSPOOL somewhere?  I guess that it should be:
/var/spool/hyla, since that has a directory called bin.
I defined it, but we bombed the same way.

> there should be a program
>called ps2fax, 
It is there.

I will try having it make the config, but here is my log:
Oct 23 16:45:12 detel FaxGetty[323]: MODEM ROCKWELL AC/V34/V1.100
Z201-V34_DP
Oct 23 16:48:07 detel login: ROOT LOGIN ON ttyp0 FROM edinct.tiac.net
Oct 23 16:52:44 detel FaxQueuer[225]: SUBMIT JOB 6
Oct 23 16:52:47 detel FaxQueuer[5091]: JOB 6: CONVERT DOCUMENT: exit
status 0xff00
Oct 23 16:52:47 detel FaxQueuer[225]: NOTIFY: bin/notify "doneq/q6"
"no_formatter" ""

  --  Ed Weinberg,
      Detel, Inc., An Internet Presence Provider
      edw@detel.com
      www.detel.com/
      www.serverking.com
      www.q5.com/  <-- find someone to CoolTalk or chat with here

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: edw@detel.com (Ed Weinberg)
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
Cc: flexfax@sgi.com
Subject: Re: flexfax: RPM: document conversion script was not found
Date: Thu, 23 Oct 1997 21:24:37 GMT
Reply-To: edw@detel.com
Sender: owner-flexfax@celestial.com

On Thu, 23 Oct 1997 16:30:12 -0400 (EDT), Nico Garcia
<raoul@cirl.meei.harvard.edu> wrote:

>OK. Pre-process it: use something like textfmt to turn the text into
>Postscript by hand, then turn the Postscript into tiffg3 with ps2fax and
>see where things break....
I have a ps2fax.gs and a few others, but no ps2fax. could that be my
probem?

  --  Ed Weinberg,
      Detel, Inc., An Internet Presence Provider
      edw@detel.com
      www.detel.com/
      www.serverking.com
      www.q5.com/  <-- find someone to CoolTalk or chat with here

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: edw@detel.com (Ed Weinberg)
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
Cc: flexfax@sgi.com
Subject: Re: flexfax: RPM: document conversion script was not found
Date: Thu, 23 Oct 1997 21:33:37 GMT
Reply-To: edw@detel.com
Sender: owner-flexfax@celestial.com

On Thu, 23 Oct 1997 16:30:12 -0400 (EDT), Nico Garcia
<raoul@cirl.meei.harvard.edu> wrote:

>OK. Pre-process it: use something like textfmt to turn the text into
>Postscript by hand, then turn the Postscript into tiffg3 with ps2fax and
>see where things break....

Got it!  I made a link:
ln -s ps2fax.gs ps2fax

Now it works.  Next I need to figure out how to get rid of that cover
page!  Thanks for everyone's help!

  --  Ed Weinberg,
      Detel, Inc., An Internet Presence Provider
      edw@detel.com
      www.detel.com/
      www.serverking.com
      www.q5.com/  <-- find someone to CoolTalk or chat with here

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Steve Williams <steve@genie96.com>
Subject: Re: flexfax: RPM: document conversion script was not found
To: edw@detel.com
Date: Thu, 23 Oct 1997 16:16:18 -0600 (MDT)
Cc: flexfax@sgi.com
Read-Receipt-To: steve@genie96.com
Sender: owner-flexfax@celestial.com

> 
> On Thu, 23 Oct 1997 13:56:26 -0600 (MDT), Steve Williams
> <steve@genie96.com> wrote:
> 
> >> >that you put something like 1 or 2 or 2.0 in that line.
> >> It came that way.  I just commented it out with a "#", rebooted (did I
> >> need to reboot?), then used the command:
> >> sendfax -d 912033674005 /etc/hosts
> >> In the hope that it would fax the text document to 9+1+203-367-4005.
> >> It bombed.
> 
> >That is VERY important that there is one less line in the messages file.
> >That means that deleteing the line fixed that problem.
> I guessed that.
> 
> >I would need to see the logfile ( I delete all my mail after reading <g> )
> >but the problem is that in your $FAXSPOOL/bin
> Should I define $FAXSPOOL somewhere?  I guess that it should be:
> /var/spool/hyla, since that has a directory called bin.
> I defined it, but we bombed the same way.
> 
> > there should be a program
> >called ps2fax, 
> It is there.
> 
> I will try having it make the config, but here is my log:
> Oct 23 16:45:12 detel FaxGetty[323]: MODEM ROCKWELL AC/V34/V1.100
> Z201-V34_DP
> Oct 23 16:48:07 detel login: ROOT LOGIN ON ttyp0 FROM edinct.tiac.net
> Oct 23 16:52:44 detel FaxQueuer[225]: SUBMIT JOB 6
> Oct 23 16:52:47 detel FaxQueuer[5091]: JOB 6: CONVERT DOCUMENT: exit
> status 0xff00
> Oct 23 16:52:47 detel FaxQueuer[225]: NOTIFY: bin/notify "doneq/q6"
> "no_formatter" ""
> 
> 

This means that for some reason, the "formatter", notably ps2fax is exiting
with a value that is non-zero.  Thus ps2fax is dying.

Try running faxsetup.


-- 
	Steve Williams, Calgary, Alberta, Canada
	Genie Computer Systems Inc.
	steve@genie96.com

"A man doesn't begin to attain wisdom until he recognizes that he is 
 no longer indespensable."
- Admiral Richard E. Byrd ( 1888-1957 )

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 19:00:55 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Ed Weinberg <edw@detel.com>
Cc: flexfax@sgi.com
Subject: Re: flexfax: RPM: document conversion script was not found
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Thu, 23 Oct 1997, Ed Weinberg wrote:

> On Thu, 23 Oct 1997 16:30:12 -0400 (EDT), Nico Garcia
> <raoul@cirl.meei.harvard.edu> wrote:
> 
> >OK. Pre-process it: use something like textfmt to turn the text into
> >Postscript by hand, then turn the Postscript into tiffg3 with ps2fax and
> >see where things break....
> I have a ps2fax.gs and a few others, but no ps2fax. could that be my
> probem?

Give the man a kewpie doll! Yes, you need to run "faxsetup" to install
a symbolic link from ps2fax.gs to ps2fax.


			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNE/XID/+ItycgIJRAQFOVAP7B9LXd7jAtW7UsgMZ3bIAznjSNeR+/J9S
Bzt8wQM4a56Ex/k7bRsv+NzkpixJAgScwZPVKUcZP55RnJx6VmfdAuYR9nx2bYQ/
x10OSmaGjWo6eX0AHUL4ADkxqk0QWyBdSL5c0RJIhoGieQETJR664r9x6gzKs3Jv
Z63/JVcHGDw=
=iyyi
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 19:05:12 -0400
From: Ramana Juvvadi <juvvadi@lekha.org>
To: Steve Williams <steve@genie96.com>
Cc: edw@detel.com, flexfax@sgi.com
Subject: Re: flexfax: RPM: document conversion script was not found
Sender: owner-flexfax@celestial.com

Steve Williams writes:

 > called ps2fax, that should fire up ghostscript.  Either it's missing,
 > or what it is doing is wrong.  Try to run this program by hand.  There is
 > a man page ( I believe ) for ps2fax ( NOTE I believe you have to be in the 
 > $FAXSPOOL directory when you run ps2fax )

Actually, this was an error in my rpm. I should have included
a symbolic link ps2fax --> ps2fax.gs in /var/spool/hyla/bin.
I will fix it in the next release of the rpm.

The ModemClass: 2 was also introduced by me in the rpm. It will be
removed in the next release. As a clueless user I typed 2 when 
faxsetup asked for ModemClass. Basically I copied the config
file faxsetup generated for me. 

I have no complaints about faxsetup. However, the information
that  faxsetup probes and finds out doesn't vary across
Readhat Linux systems. It is the information that faxaddmodem
is trying to find out that is likely to vary from 
machine to machine. Unfortuantely, that is the information
that is most difficult to find out by probing. I don't
mean to complain about the program as such. I just think
that it is difficult for any program to assure that
probing a modem works reliably. I don't see how
a programmer can test his program on hundreds of 
models. 

Ramana

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Steve Williams <steve@genie96.com>
Subject: Re: flexfax: RPM: document conversion script was not found
To: juvvadi@lekha.org (Ramana Juvvadi)
Date: Thu, 23 Oct 1997 17:21:43 -0600 (MDT)
Cc: steve@genie96.com, edw@detel.com, flexfax@sgi.com
Read-Receipt-To: steve@genie96.com
Sender: owner-flexfax@celestial.com

> 
> Steve Williams writes:
> 
>  > called ps2fax, that should fire up ghostscript.  Either it's missing,
>  > or what it is doing is wrong.  Try to run this program by hand.  There is
>  > a man page ( I believe ) for ps2fax ( NOTE I believe you have to be in the 
>  > $FAXSPOOL directory when you run ps2fax )
> 
> Actually, this was an error in my rpm. I should have included
> a symbolic link ps2fax --> ps2fax.gs in /var/spool/hyla/bin.
> I will fix it in the next release of the rpm.
> 
> The ModemClass: 2 was also introduced by me in the rpm. It will be
> removed in the next release. As a clueless user I typed 2 when 
> faxsetup asked for ModemClass. Basically I copied the config
> file faxsetup generated for me. 
> 
> I have no complaints about faxsetup. However, the information
> that  faxsetup probes and finds out doesn't vary across
> Readhat Linux systems. It is the information that faxaddmodem

Ummm, it might have caught the problem with ps2fax, because I know it 
checks for the PS RIP, and makes the appropriate link.


> is trying to find out that is likely to vary from 
> machine to machine. Unfortuantely, that is the information
> that is most difficult to find out by probing. I don't
> mean to complain about the program as such. I just think
> that it is difficult for any program to assure that
> probing a modem works reliably. I don't see how
> a programmer can test his program on hundreds of 
> models. 
> 
> Ramana
> 
> 


-- 
	Steve Williams, Calgary, Alberta, Canada
	Genie Computer Systems Inc.
	steve@genie96.com

"A man doesn't begin to attain wisdom until he recognizes that he is 
 no longer indespensable."
- Admiral Richard E. Byrd ( 1888-1957 )

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 20:11:59 -0400
From: Ramana Juvvadi <juvvadi@lekha.org>
To: Steve Williams <steve@genie96.com>
Cc: juvvadi@lekha.org (Ramana Juvvadi), edw@detel.com, flexfax@sgi.com
Subject: Re: flexfax: RPM: document conversion script was not found
Sender: owner-flexfax@celestial.com

Steve Williams writes:

 > 
 > Ummm, it might have caught the problem with ps2fax, because I know it 
 > checks for the PS RIP, and makes the appropriate link.
 > 

Yes, but it doesn't take long for the rpm to reach a 
stable release especially with he help of beta-testers 
like Ed Weinberg -:)

Ramana

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthew.Rice@ftlsol.com
X-Authentication-Warning: hudson.ftlsol.com: matt set sender to Matthew.Rice using -f
Subject: Re: flexfax: RPM: document conversion script was not found
To: edw@detel.com
Date: Thu, 23 Oct 1997 21:11:10 -0400 (EDT)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> Now it works.  Next I need to figure out how to get rid of that cover
> page!  

man sendfax.  It's all there.

BTW, looks like we could use a flexfax-rpm list :)

-- 
Matthew Rice                             e-mail: matthew.rice@ftlsol.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: RPM: document conversion script was not found
To: juvvadi@lekha.org (Ramana Juvvadi)
Date: Fri, 24 Oct 1997 08:03:12 +0200 (MSZ)
Cc: steve@genie96.com, edw@detel.com, flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Ramana Juvvadi wrote:
    
    Steve Williams writes:
    
     > called ps2fax, that should fire up ghostscript.  Either it's missing,
     > or what it is doing is wrong.  Try to run this program by hand.  There is
     > a man page ( I believe ) for ps2fax ( NOTE I believe you have to be in the 
     > $FAXSPOOL directory when you run ps2fax )
    
    Actually, this was an error in my rpm. I should have included
    a symbolic link ps2fax --> ps2fax.gs in /var/spool/hyla/bin.
    I will fix it in the next release of the rpm.

Once again: the ERROR isn't the missing symlink in your
package, the REAL ERROR is not using faxsetup(1M) after
installing the binaries; BTW: there also hooks in
faxsetup(1M) if you have to do special tasks on some
UNIX types.
    
    The ModemClass: 2 was also introduced by me in the rpm. It will be
    removed in the next release. As a clueless user I typed 2 when 
    faxsetup asked for ModemClass. Basically I copied the config
    file faxsetup generated for me. 
    
    I have no complaints about faxsetup. However, the information
    that  faxsetup probes and finds out doesn't vary across
    Readhat Linux systems. It is the information that faxaddmodem
    is trying to find out that is likely to vary from 
    machine to machine. Unfortuantely, that is the information
    that is most difficult to find out by probing. I don't
    mean to complain about the program as such. I just think
    that it is difficult for any program to assure that
    probing a modem works reliably. I don't see how
    a programmer can test his program on hundreds of 
    models. 
    
faxaddmodem(1M) takes the RS-232 specified on the command
line or prompted for and tries to test the modem and check
if the modem is ready for use (e.g. answers to AT-cmds),
checks the modem model (to use a prototype config file)
and tries to create a modem config file exactly for this
type of modem. If the modem doesn't work with faxaddmodem(1M)
there is less hope that it will work with HylaFAX.
If faxaddmodem(1M) doesn't work on RH Linux one should fix
this script. PERIOD.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: edw@detel.com (Ed Weinberg)
To: flexfax@sgi.com
Subject: flexfax: Installing new RPM...
Date: Mon, 20 Oct 1997 19:00:58 GMT
Reply-To: edw@detel.com
Sender: owner-flexfax@celestial.com

I just installed the new RPM for hylafax-4.0-3.

I re-booted and when I try to use sendfax:
# sendfax -d 12033674005 /etc/hosts
(I hope this sends the file /etc/hosts to 1-203-367-4005)

I get the following message:
Font Courier-Bold: /usr/lib/hyla/afm/Courier-Bold: Can not open font
metrics file; using fixed widths.
460 Failed to submit job 2: Unable to open scheduler FIFO: No such
device or address.

Any ideas what is wrong?

  --  Ed Weinberg,
      Detel, Inc., An Internet Presence Provider
      edw@detel.com
      www.detel.com/
      www.serverking.com
      www.q5.com/  <-- find someone to CoolTalk or chat with here

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: evan@telly.org (Evan Leibovitch)
Subject: Re: flexfax: Installing new RPM...
To: edw@detel.com
Date: Mon, 20 Oct 1997 22:49:40 -0400 (EDT)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

Ed Weinberg writes:

> I just installed the new RPM for hylafax-4.0-3.
 
> I get the following message:
> Font Courier-Bold: /usr/lib/hyla/afm/Courier-Bold: Can not open font
> metrics file; using fixed widths.
> 460 Failed to submit job 2: Unable to open scheduler FIFO: No such
> device or address.
> 
> Any ideas what is wrong?

I haven't tried the RPM, but if your Linux system is like mine
HylaFax's 'textfmt' doesn't llke the way the new Ghostscript
maintains its font-metric files. The new one uses a filename
called "Fontmap" to map fontnames to filenames, in the old version
the font name *was* the file name.

I've uploaded a perl script to Sam that creates the symbolic links
necessary to make textfmt happy.

-- 
  Evan Leibovitch, Sound Software Ltd, located in beautiful Brampton, Ontario
 Supporting PC-based Unix since 1985 / Caldera & SCO authorized / www.telly.org
When the only tool you have is a hammer, every problem comes to look like a nail

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 21 Oct 1997 13:44:36 +0200
To: flexfax@sgi.com
From: Ask Bjrn Hansen <ask@netcetera.dk>
Subject: Re: flexfax: Installing new RPM...
Sender: owner-flexfax@celestial.com

At 21:00 +0200 20/10/97, Ed Weinberg wrote:
>I just installed the new RPM for hylafax-4.0-3.

>I get the following message:
>Font Courier-Bold: /usr/lib/hyla/afm/Courier-Bold: Can not open font
>metrics file; using fixed widths.

Get the .src.rpm version and change the .spec file to "--with-AFM=yes" and
rebuild..


kind regards,

ask

---------------------------------------------------------------------
ask bjoern hansen - Netcetera - Finsensvej 80 - DK-2000 Frederiksberg
tlf 38 88 32 22 / 40 44 58 66 / 38 88 20 38 ext 341 - Fax 38 88 30 38
Webdesign, Webhotel, Mailhotel, UUCP & more! http://www.netcetera.dk/

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: edw@detel.com (Ed Weinberg)
To: Ask Bjrn Hansen <ask@netcetera.dk>
Cc: flexfax@sgi.com
Subject: Re: flexfax: Installing new RPM...
Date: Tue, 21 Oct 1997 13:42:05 GMT
Reply-To: edw@detel.com
Sender: owner-flexfax@celestial.com

On Tue, 21 Oct 1997 13:44:36 +0200, Ask Bjrn Hansen
<ask@netcetera.dk> wrote:

>At 21:00 +0200 20/10/97, Ed Weinberg wrote:
>>I just installed the new RPM for hylafax-4.0-3.
>
>>I get the following message:
>>Font Courier-Bold: /usr/lib/hyla/afm/Courier-Bold: Can not open font
>>metrics file; using fixed widths.
>
>Get the .src.rpm version and change the .spec file to "--with-AFM=yes" and
>rebuild..
I have never changed an rpm src.  Do you mean a file called ".spec"?  
Will the source come with instructions to rebuild, or is this
something I can find in the rpm man page?

What does this change do?

What is different with my RedHat instalation which requires this
change.
  --  Ed Weinberg,
      Detel, Inc., An Internet Presence Provider
      edw@detel.com
      www.detel.com/
      www.serverking.com
      www.q5.com/  <-- find someone to CoolTalk or chat with here

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 21 Oct 1997 15:56:46 +0200
To: edw@detel.com
From: Ask Bjrn Hansen <ask@netcetera.dk>
Subject: Re: flexfax: Installing new RPM...
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

>On Tue, 21 Oct 1997 13:44:36 +0200, Ask Bjrn Hansen
><ask@netcetera.dk> wrote:
>
>>At 21:00 +0200 20/10/97, Ed Weinberg wrote:
>>>I just installed the new RPM for hylafax-4.0-3.
>>
>>>I get the following message:
>>>Font Courier-Bold: /usr/lib/hyla/afm/Courier-Bold: Can not open font
>>>metrics file; using fixed widths.
>>
>>Get the .src.rpm version and change the .spec file to "--with-AFM=yes" and
>>rebuild..
>I have never changed an rpm src.  Do you mean a file called ".spec"?

Yup.

>Will the source come with instructions to rebuild, or is this
>something I can find in the rpm man page?

Check the RPM HOWTO at www.rpm.org.

Something like:

rpm -U hylafax-4.0-3.src.rpm
<youreditor> /usr/src/redhat/spec/hylafax-4.0-3.spec
edit the AFM line
rpm -ba /usr/src/redhat/spec/hylafax-4.0-3.spec
rpm -U /usr/src/redhat/RPMS/i386/hylafax-4.0-3.i386.rpm

should do it..

>What does this change do?

Compile (and/or install..) hylafax with 'AFM'.

>What is different with my RedHat instalation which requires this
>change.

Nothing I guess; I had to do the same thing. :-)


regards,

ask

---------------------------------------------------------------------
ask bjoern hansen - Netcetera - Finsensvej 80 - DK-2000 Frederiksberg
tlf 38 88 32 22 / 40 44 58 66 / 38 88 20 38 ext 341 - Fax 38 88 30 38
Webdesign, Webhotel, Mailhotel, UUCP & more! http://www.netcetera.dk/

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: edw@detel.com (Ed Weinberg)
To: Ask Bjrn Hansen <ask@netcetera.dk>
Cc: flexfax@sgi.com
Subject: Re: flexfax: Installing new RPM...
Date: Tue, 21 Oct 1997 17:14:59 GMT
Reply-To: edw@detel.com
Sender: owner-flexfax@celestial.com

First, I very much appreciate your help.

You mentioned you had to do this too, but from reading this newsgroup
it was my assumption that the RPM would do everything for me!

Next...
There were a few differences like the directory spec was SPEC, but
other than that I was able to make the changes.  when I ran:

On Tue, 21 Oct 1997 15:56:46 +0200, Ask Bjrn Hansen
<ask@netcetera.dk> wrote:

>rpm -ba /usr/src/redhat/spec/hylafax-4.0-3.spec
(actually "rpm -ba /usr/src/redhat/SPECS/hylafax-4.0.spec")
It ran through a list of many file paths then I got the following
response:
+ [ 0 -ne 0 ]
+ cd hylafax-v4.0pl1
+ cd /usr/src/redhat/BUILD/hylafax-v4.0pl1
+ chown -R root.root .
+ chmod -R a+rX,g-w,o-w .
+ echo Patch #0:
Patch #0:
+ patch -p1 -s
+ exit 0
+ umask 022
+ echo Executing: %build
Executing: %build
+ cd /usr/src/redhat/BUILD
+ cd hylafax-v4.0pl1
+ configure --with-DIR_BIN=/usr/bin --with-DIR_SBIN=/usr/sbin
--with-DIR_LIBEXEs
/var/tmp/rpmbu08561baa: configure: command not found
Bad exit status
(end of response)

What did I do wrong?
Is there a specific directory I should have run this in?
My guess is that the command is supposed to build me a new RPM to
install.

Thanks again.

  --  Ed Weinberg,
      Detel, Inc., An Internet Presence Provider
      edw@detel.com
      www.detel.com/
      www.serverking.com
      www.q5.com/  <-- find someone to CoolTalk or chat with here

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 03:13:30 -0400 (EDT)
From: Dave Wreski <dave@nic.com>
To: edw@detel.com
Subject: Re: flexfax: Installing new RPM...
Cc: flexfax@sgi.com, Ask=?us-ascii?Q?_Bj=F8rn?= Hansen <ask@netcetera.dk>
Sender: owner-flexfax@celestial.com

> Executing: %build
> + cd /usr/src/redhat/BUILD
> + cd hylafax-v4.0pl1
> + configure --with-DIR_BIN=/usr/bin --with-DIR_SBIN=/usr/sbin
> --with-DIR_LIBEXEs
> /var/tmp/rpmbu08561baa: configure: command not found
> Bad exit status
> (end of response)

Before you run it, put your current directory in your path:

bash# export PATH=$PATH:.

Then try it again.

> What did I do wrong?
> Is there a specific directory I should have run this in?
> My guess is that the command is supposed to build me a new RPM to
> install.

Correct, but it doesn't matter where you run rpm from, as long as you use
absolute paths to the directory the source is in.  After it actually compiles,
your done. At that point it _is_ installed.

Also, be sure you actually have the C compiler, gcc, installed.  Hopefully you
have the CDROM because it will probably require a package or two, and gcc
itself is pretty large..

Dave

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 21 Oct 1997 14:38:22 -0400
From: Ramana Juvvadi <juvvadi@lekha.org>
To: flexfax@sgi.com
Subject: flexfax: Announcement: RPMs for hylafax and libgr  
Sender: owner-flexfax@celestial.com


Since the last announcement, I have gone through some more builds.
The results can be dowloaded from

	ftp://ftp.lekha.org/pub
	http://www.lekha.org/hyalfax


Hylafax rpm has advanced to release 4 now, whereas libgr has advanced
to release 2. 

I'll follow up with a mail on some of the  the questions that were raised.

I sent the following announcement to comp.os.linux.announce and I put
the rpms in the incoming directory of ftp.redhat.com

-------------------------------------------------------------------


What is hylafax ?
-----------------

Hylafax is a package for sending and receiving faxes. Hylafax is an 
enterprise soln. It can spool the outgoing and incoming faxes on
a central server. The client can be either on a unix or a Windows platform.


About the RPM of the hylafax
-----------------------------

For the impatient, the URLs sre

	http://www.lekha.org/hylafax
	ftp://ftp.lekha.org/pub


Hylafax is a great package, but it doesn't work out of the box. It
requires an elaborate setup. My aim in building this package is to
reduce some of this complexity for Redhat Linux. The only thing you to
do after the installing the RPM is to edit a config.modem file to suit
the parameters of your modem.


There is already a linux binary at SGI's site. An rpm for hylafax was
built using this binary by Matthew Rice before.  Some of you who tried
to install the rpm might have experienced problems running
'faxsetup'. This is because the rpm doesn't have all the files.  Also,
it doesn't set up the package for you.  After encouragement from
Matthew, I decided to build a new rpm of hylafax from the source.

The features of the new rpm are: 

   1.	Redhat Linux 4.2 comes with libgr-2.0.9. However, the libtiff
	that comes in libgr-2.0.9 is older than the version required by
	hylafax. Meanwhile libgr has evolved and the latest version is
	libgr-2.0.13. However,libgr-2.0.13 has some shell scripts missing. I
	added them to make it libgr-2.0.14.

   2.	libgr-2.0.14 subsumes the functionality of another package 
	that comes with Redhat 4.2, netpbm-1mar1994. 

   3.	It comes with rpms of the sources of hylafax and libgr. 

   4.   It comes with an automatic install at the time of 'rpm -i'
	and uninstall at the time of 'rpm -e'. 

   5. 	It includes scripts /etc/rc.d for automatic startup at boot time.

   6.   It includes symbolic links to ghostscript afm files for 
	conversion from ascii to postscript 
	
   7.	It uses /usr instead of /usr/local in the tradition of rpm 

   8.	It uses the directories /var/spool/hyla, /usr/lib/hyla instead of
	/var/spool/fax, /usr/lib/fax. The reason is that there are two 
	other fax pages for linux, efax and mgetty+sendfax. I feel it 
	is better to have a unique name like
	hyla than a generic one like fax causing mixup 

   9.	The documentation directory is set to /usr/doc instead of /var/httpd 

   10.	There is an executable called xferstats in hylafax clashing with 
	an xferstats from wu-ftpd. Hence its name is changed to 
	xfersfaxtats. Similarly xferlog is changed to  xferfaxlog. 


	http://www.lekha.org/hylafax
	ftp://ftp.lekha.org/pub

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: edw@detel.com (Ed Weinberg)
To: Ramana Juvvadi <juvvadi@lekha.org>
Cc: flexfax@sgi.com
Subject: Re: flexfax: Announcement: RPMs for hylafax and libgr  
Date: Tue, 21 Oct 1997 21:14:48 GMT
Reply-To: edw@detel.com
Sender: owner-flexfax@celestial.com

On Tue, 21 Oct 1997 16:35:11 -0400, Ramana Juvvadi <juvvadi@lekha.org>
replied to me:

> > Does this or any other packate have to be uninstalled before the new
> > package is installed?
>
>Yes, netpbm-mar... has to be uninstalled. I saw your message in
>flexfax mailing list. Upgrade to ghostscript-5. I posted 
>a message in the mailing list also.

netpbm-mar...what? I have no netpbm-mar is not installed in my
computer.  Is there an RPM for ghostscript-5?  I currently have RPM
ghostscript-3.33-2 installed.

  --  Ed Weinberg,
      Detel, Inc., An Internet Presence Provider
      edw@detel.com
      www.detel.com/
      www.serverking.com
      www.q5.com/  <-- find someone to CoolTalk or chat with here

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Announcement: RPMs for hylafax and libgr
To: juvvadi@lekha.org (Ramana Juvvadi)
Date: Wed, 22 Oct 1997 14:14:09 +0200 (MSZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Ramana Juvvadi wrote:
    
    
    Since the last announcement, I have gone through some more builds.
    The results can be dowloaded from
    
    	ftp://ftp.lekha.org/pub
    	http://www.lekha.org/hyalfax

[ BTW: It's a good idea to send also the MD5 checksums for your
binary distribution so that folks can check what they transfered. ]
    
Thanks for your work and don't get my comments below wrong.

HylaFAX (please note the spelling of the name) is a free
software project and everyone is free to do whatever he/she
wants to do with it. It is supported by folks of this list and
by the keepers of binary distributions (you're now one of
them). It's difficult for us to help other people out if
they don't use the tools (or don't have these tools in their
binary distribution), e.g. faxsetup(1M) or faxaddmodem(1M),
or even if the man pages which are part of the source distribution
don't match the names of executables (don't know if you changed
also all man and HTML pages for the changed name(s) of commands).

    Hylafax rpm has advanced to release 4 now, whereas libgr has advanced
    to release 2. 

The current version of HylaFAX is v4.0pl1. It would be good if
the RPM binary would have the same name in it and (if needed)
some additional version information added. It should be clear
that the distributed binaries are compiled from v4.0pl1 (we're
moving to v4.0pl2 sometimes).
    
    I'll follow up with a mail on some of the  the questions that were raised.
    
    I sent the following announcement to comp.os.linux.announce and I put
    the rpms in the incoming directory of ftp.redhat.com
    
    -------------------------------------------------------------------
    
    
    What is hylafax ?
    -----------------
    
    Hylafax is a package for sending and receiving faxes. Hylafax is an 
    enterprise soln. It can spool the outgoing and incoming faxes on
    a central server. The client can be either on a unix or a Windows platform.

There are also clients for Mac's.
    
    About the RPM of the hylafax
    -----------------------------
    
    For the impatient, the URLs sre
    
    	http://www.lekha.org/hylafax
    	ftp://ftp.lekha.org/pub
    
    
    Hylafax is a great package, but it doesn't work out of the box. It
    requires an elaborate setup. My aim in building this package is to
    reduce some of this complexity for Redhat Linux. The only thing you to
    do after the installing the RPM is to edit a config.modem file to suit
    the parameters of your modem.

Binary distributions are a good thing. This allow folks without
the tools to compile and build from source (e.g. gcc compiler)
and without the appropriate knowlege to make use of this software.

The HylaFAX source and all other binary distributions come with
a lot of tools to make the configuration part of the installation
as easy as possible. These tools (e.g. faxsetup(1M)) check your
system and (not only but also) the installed HylaFAX software
for "is it ready to use?". They let you create needed files or information
(e.g. entries in /etc/services) and so on. faxaddmodem(1M)
will *test* your modem and try to find and build a config file 
for it. A modem which causes problems during faxaddmodem(1M)
will perhaps not work together with HylaFAX. Creating modem
config files by hand is possible but needs much more experience
with HylaFAX and the modems as one needs for using faxaddmodem(1M).

May be you don't get my points. But we will all see these
questions in the mailing list in the future like this one
just arriving:

> To: flexfax@sgi.com
> Subject: flexfax: Newest build give this error...
> 
> I installed the latest hylafax RPM.  How do I specify the modem?
                                       ^^^^^^^^^^^^^^^^^^^^^^^^^^
> I re-booted.
> I tried to use the command sendfax and got this error:

	....
    
    
    There is already a linux binary at SGI's site. An rpm for hylafax was
    built using this binary by Matthew Rice before.  Some of you who tried
    to install the rpm might have experienced problems running
    'faxsetup'. This is because the rpm doesn't have all the files.  Also,
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

This is definitely not a good idea, I think, sorry.

	matthias

    it doesn't set up the package for you.  After encouragement from
    Matthew, I decided to build a new rpm of hylafax from the source.
    
    The features of the new rpm are: 
    
	....

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 10:01:52 -0400
From: Ramana Juvvadi <juvvadi@lekha.org>
To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
Cc: juvvadi@lekha.org (Ramana Juvvadi), flexfax@sgi.com
Subject: Re: flexfax: Announcement: RPMs for hylafax and libgr
Sender: owner-flexfax@celestial.com

Matthias.Apitz@SOFTCON.de writes:

 > [ BTW: It's a good idea to send also the MD5 checksums for your
 > binary distribution so that folks can check what they transfered. ]

Sure. Would check into it.

 > HylaFAX (please note the spelling of the name) is a free

I have no problem sticking to the spelling. The source code on ftp.sgi.com
says hylafax-v4.0pl1.tar.gz. Is it possible for you to change its
name, because I have to refer to the original URL in RPM spec file.


 > The HylaFAX source and all other binary distributions come with
 > a lot of tools to make the configuration part of the installation
 > as easy as possible. These tools (e.g. faxsetup(1M)) check your
 > system and (not only but also) the installed HylaFAX software
 > for "is it ready to use?". They let you create needed files or information
 > (e.g. entries in /etc/services) and so on. faxaddmodem(1M)
 > will *test* your modem and try to find and build a config file 
 > for it. A modem which causes problems during faxaddmodem(1M)
 > will perhaps not work together with HylaFAX. Creating modem
 > config files by hand is possible but needs much more experience
 > with HylaFAX and the modems as one needs for using faxaddmodem(1M).


I'll follow up with a mail on the 'faxaddmodem' issue.



 >     There is already a linux binary at SGI's site. An rpm for hylafax was
 >     built using this binary by Matthew Rice before.  Some of you who tried
 >     to install the rpm might have experienced problems running
 >     'faxsetup'. This is because the rpm doesn't have all the files.  Also,
 >                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 > 
 > This is definitely not a good idea, I think, sorry.
 > 
 > 	matthias

I meant the old rpm that was built by Matthew Rice didn't have all the
files. That was also one of the reasons I decided to build a new RPM.

Ramana

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthew.Rice@ftlsol.com
X-Authentication-Warning: hudson.ftlsol.com: matt set sender to Matthew.Rice using -f
Subject: Re: flexfax: Announcement: RPMs for hylafax and libgr
To: juvvadi@lekha.org (Ramana Juvvadi)
Date: Wed, 22 Oct 1997 14:24:31 -0400 (EDT)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

[lurk mode OFF]

Quite a fun little thread we have going here...
To answer a few questions that I've seen popping up.

1) I did inform Ramana about the upcoming patch (ya gotta admire his
   keenness).

2) Is there any reason that we cannot rename the xferstat/xferlog for
   the next release?
  
3) Why not install the documentation in /usr/doc and add links to
   them in /var/httpd/... (or vice versa)?

4) I don't mind /usr/... as the install directory.  Although, I can see
   one good reason why a newbie would like /usr/local/... :

       matt@hudson:~/src[58]% ls /usr/bin |wc -l
          1006
       matt@hudson:~/src[59]% ls /usr/local/bin |wc -l
            55

5) I believe that Ramana did keep faxaddmodem and faxsetup in the package.

 
> I have no problem sticking to the spelling. The source code on ftp.sgi.com
> says hylafax-v4.0pl1.tar.gz. Is it possible for you to change its
> name, because I have to refer to the original URL in RPM spec file.

My intention was to let Ramana's RPM become the official version.  I
will still supply a tarball for those that aren't configuratively
challenged.  My RPM wasn't anything more than a tarball with package
tracking.  I had no interest in making it anything more complex so I
welcome Ramana's efforts.


>>     'faxsetup'. This is because the rpm doesn't have all the files.  Also,
>>                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> 
> I meant the old rpm that was built by Matthew Rice didn't have all the
> files. That was also one of the reasons I decided to build a new RPM.

Seeing as the RPM is an exact (or should be) reflection of the tar'ed
distribution I would like to know what is missing.  I intentionally
left out items from other packages (tiff utils, afm files, gawk, etc).
Instead, I added references to these items in the linux specific README.

Regards,
-- 
Matthew Rice                             e-mail: matthew.rice@ftlsol.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Announcement: RPMs for hylafax and libgr
To: juvvadi@lekha.org (Ramana Juvvadi)
Date: Wed, 22 Oct 1997 22:13:24 +0200 (MSZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Ramana Juvvadi wrote:
    
    Matthias.Apitz@SOFTCON.de writes:
    
     > [ BTW: It's a good idea to send also the MD5 checksums for your
     > binary distribution so that folks can check what they transfered. ]
    
    Sure. Would check into it.
    
     > HylaFAX (please note the spelling of the name) is a free
    
    I have no problem sticking to the spelling. The source code on ftp.sgi.com
    says hylafax-v4.0pl1.tar.gz. Is it possible for you to change its
    name, because I have to refer to the original URL in RPM spec file.

The spelling of HylaFAX in URLs should not be changed. I was talking
about the spelling in the documentation and in the mailing-list.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Announcement: RPMs for hylafax and libgr
To: Matthew.Rice@ftlsol.com
Date: Thu, 23 Oct 1997 08:18:45 +0200 (MSZ)
Cc: juvvadi@lekha.org (Ramana Juvvadi), flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Matthew.Rice@ftlsol.com wrote:
    
    <...stuff deleted...>
    
    2) Is there any reason that we cannot rename the xferstat/xferlog for
       the next release?

Yes there are reasons not to rename the names of xferstats(1M)
and xferlog(4): the existing installations all over the world,
the crontab scripts or other scripts on servers written by
sysadmins, the need of changing man pages, HTML, FAQs etc.
And I can see no problem with that names. If they are clashing
with other (sysadmin!) tools just install them in places 
where they do not owerwrite existing files.

There will be no change for this names in v4.0pl2 source tree.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 11:41:00 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Super-User <root@peach.asiainter.net>
Cc: Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: Announcement: RPMs for hylafax and libgr
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Thu, 23 Oct 1997, Super-User wrote:

> On Wed, 22 Oct 1997 Matthew.Rice@ftlsol.com wrote:
> > 2) Is there any reason that we cannot rename the xferstat/xferlog for
> >    the next release?
> 
> yes, it makes more work for everyone to keep track of everything, and
> provides no benefit.
> why did microsoft change index.htm[l] to default.htm?
> in microsoft internet mail, why do you separate multiple recipients with
> semicolons instead of commas?

Because Microsnot likes to break compatibility, forcing people to their
products.

> stupid little changes like this just suck.  there are loads of other
> little changes like this in various software that are too numerous to
> count.
> many commands between solaris 1.x and 2.x are different, but that's
> understandable because it's a different product entirely..
> 
> my 2 cents - dont change something unless there's a benefit.

In this case, there is a specific benefit: xferstats in the Linux
package is already present in the wu-ftpd binaries, and they will
overwrite each other in /usr/sbin in configurations done consistently
with RedHat's model of where such things go. It doesn't traumatically
affect other OS's: it makes sense.

There have been similar suggestions in the past: renaming the received
faxes to "filename.fax" instead of "filename.tif" to ease Web service
of received documents, for example. I don't think that one took
off, but I modified faxrcvd appropriately on *my* installation.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNE9wAj/+ItycgIJRAQEY+AP7Bm+8M07XVuwa3mSrvRUs8JoIjmAaA4OD
O3+DjLg758ro1n8uW1ziFRCyBgGoDemrLxlTc7Kde45yOvTB2qX0jr0HOFX94PET
eQ48uybqtJVdQFDHLfSyxk5G0Xe1x2FW1DKlPqBEIjrMEsfXwkzO5nTwucml7XGp
OhG/sksdca0=
=WMIn
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 14:28:52 -0500 (CDT)
From: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
To: Super-User <root@peach.asiainter.net>
cc: Matthew.Rice@ftlsol.com, Ramana Juvvadi <juvvadi@lekha.org>,
        flexfax@sgi.com
Subject: Re: flexfax: Announcement: RPMs for hylafax and libgr
Sender: owner-flexfax@celestial.com

I fully agree. Another thing to remember, another software version to
create if you are doing anything with the xferstat/xferlog and need to
maintain remote stations that you don't want to install the latest patch
in, etc. etc.

Julio

On Thu, 23 Oct 1997, Super-User wrote:

> 
> On Wed, 22 Oct 1997 Matthew.Rice@ftlsol.com wrote:
> > 2) Is there any reason that we cannot rename the xferstat/xferlog for
> >    the next release?
> 
> yes, it makes more work for everyone to keep track of everything, and
> provides no benefit.
> why did microsoft change index.htm[l] to default.htm?
> in microsoft internet mail, why do you separate multiple recipients with
> semicolons instead of commas?
> 
> stupid little changes like this just suck.  there are loads of other
> little changes like this in various software that are too numerous to
> count.
> many commands between solaris 1.x and 2.x are different, but that's
> understandable because it's a different product entirely..
> 
> my 2 cents - dont change something unless there's a benefit.
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 29 Oct 1997 08:44:41 -0500 (EST)
From: Evan Leibovitch <evan@telly.org>
To: Super-User <root@peach.asiainter.net>
cc: Matthew.Rice@ftlsol.com, Ramana Juvvadi <juvvadi@lekha.org>,
        flexfax@sgi.com
Subject: Re: flexfax: Announcement: RPMs for hylafax and libgr
Sender: owner-flexfax@celestial.com



On Thu, 23 Oct 1997, Super-User wrote:

Using the 'root' account for routine things (such as sending mail)
is not usually a wise idea.

> On Wed, 22 Oct 1997 Matthew.Rice@ftlsol.com wrote:
> > 2) Is there any reason that we cannot rename the xferstat/xferlog for
> >    the next release?
 
> yes, it makes more work for everyone to keep track of everything, and
> provides no benefit.

No benefit to you, maybe, but the case has been clearly made.

The name 'xferstat' is sufficiently vague that it can -- and has --
been duplicated by other software. This issue didn't even arise until
someone ran into problems because of the redundant filenames.

If you'd been following this thread you'd know the rationale. This
is not a gratuitous request.

- Evan

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: Announcement: RPMs for hylafax and libgr
To: evan@telly.org (Evan Leibovitch)
Date: Thu, 30 Oct 1997 08:15:30 +0000 (GMT)
Cc: root@peach.asiainter.net, Matthew.Rice@ftlsol.com, juvvadi@lekha.org,
        flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> The name 'xferstat' is sufficiently vague that it can -- and has --
> been duplicated by other software. This issue didn't even arise until
> someone ran into problems because of the redundant filenames.

Can someone explain why a conflict arises.  In the beta of Hylafax 4
that I have, all the logs are written into the /usr/spool/fax tree, so
cannot conflict with /var/adm etc.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 31 Oct 1997 09:34:39 -0500
From: Ramana Juvvadi <juvvadi@lekha.org>
To: David Woolley <david@djwhome.demon.co.uk>
Cc: evan@telly.org (Evan Leibovitch), root@peach.asiainter.net,
        Matthew.Rice@ftlsol.com, juvvadi@lekha.org, flexfax@sgi.com
Subject: Re: flexfax: Announcement: RPMs for hylafax and libgr
Sender: owner-flexfax@celestial.com

David Woolley writes:
 > > The name 'xferstat' is sufficiently vague that it can -- and has --
 > > been duplicated by other software. This issue didn't even arise until
 > > someone ran into problems because of the redundant filenames.
 > 
 > Can someone explain why a conflict arises.  In the beta of Hylafax 4
 > that I have, all the logs are written into the /usr/spool/fax tree, so
 > cannot conflict with /var/adm etc.

There is an xferstats in /usr/sbin from wu-ftpd. The xferlog clash
is with the man page. It was probably not noticed before because nobody
tried to install it in /usr.

However, the names are clobbered on even on /usr/local. Which
xferstats is called depends on whether you have /usr/local/sbin
first in your path or /usr/sbin. The same goes for xferlog man page.

Ramana

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Announcement: RPMs for hylafax and libgr
To: juvvadi@lekha.org (Ramana Juvvadi)
Date: Fri, 31 Oct 1997 18:40:10 +0100 (MEZ)
Cc: david@djwhome.demon.co.uk, evan@telly.org, root@peach.asiainter.net,
        Matthew.Rice@ftlsol.com, juvvadi@lekha.org, flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Ramana Juvvadi wrote:
    
    David Woolley writes:
     > > The name 'xferstat' is sufficiently vague that it can -- and has --
     > > been duplicated by other software. This issue didn't even arise until
     > > someone ran into problems because of the redundant filenames.
     > 
     > Can someone explain why a conflict arises.  In the beta of Hylafax 4
     > that I have, all the logs are written into the /usr/spool/fax tree, so
     > cannot conflict with /var/adm etc.
    
    There is an xferstats in /usr/sbin from wu-ftpd. The xferlog clash
    is with the man page. It was probably not noticed before because nobody
    tried to install it in /usr.
    
    However, the names are clobbered on even on /usr/local. Which
    xferstats is called depends on whether you have /usr/local/sbin
    first in your path or /usr/sbin. The same goes for xferlog man page.

xferstats(1M) is used in faxcron(1M) and perhaps mostly in scripts
like crontab(4F) or other maintainance stuff; in scripts you may
(in crontab you must) use full pathnames to start this tool
or to use the xferlog(4F) file itself; as I said I can't see much
bennefit renaming these things; on the other side renaming will
affect all existing scripts/crontabs and running installations of
HylaFAX;

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Leslie Mikesell <les@Mcs.Net>
Subject: Re: flexfax: Announcement: RPMs for hylafax and libgr
To: david@djwhome.demon.co.uk (David Woolley)
Date: Sat, 1 Nov 1997 13:04:18 -0600 (CST)
Cc: evan@telly.org, root@peach.asiainter.net, Matthew.Rice@ftlsol.com,
        juvvadi@lekha.org, flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> 
> > The name 'xferstat' is sufficiently vague that it can -- and has --
> > been duplicated by other software. This issue didn't even arise until
> > someone ran into problems because of the redundant filenames.
> 
> Can someone explain why a conflict arises.  In the beta of Hylafax 4
> that I have, all the logs are written into the /usr/spool/fax tree, so
> cannot conflict with /var/adm etc.

It has to do with the Linux Filesystem Standard which, in an attempt
to normalize the location of files across different distributions,
insists that anything included in a complete system distribution
*not* live in /usr/local but have its executables in the standard
places, config files in /etc, logs in /var/log, and so on.  It's
one of those basically good ideas that that doesn't quite match
reality...  Aside from the problem of duplicate filenames there
is also the problem that if you need an update ahead of the distributor
and compile your own you end up having to find all the changed
locations and modify your crontabs, PATH, and assorted other things
to match.

But, it's only an issue if RedHat, Debian, etc. include the progam
as part of their stock system.  Third party add-ons are expected
to live in /usr/local as they would if you compiled your own.
   
 Les Mikesell
     les@mcs.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 21 Oct 1997 15:31:16 -0400
From: Ramana Juvvadi <juvvadi@lekha.org>
To: flexfax@sgi.com
Subject: flexfax: Hylafax rpm issues
Sender: owner-flexfax@celestial.com


I hope with this rpm release, hylafax is easier than before to setup.

The old annoyances of afm directories, entries in /etc/aliases, 
/etc/inetd.conf, /etc/inittab, /etc/services, creating of the user
fax are all taken care of.

To send and receive faxes all one needs to do is

	(1) rpm -i hylafax-4.0-4.i386.rpm
	(2) reboot
	(3) Edit /var/spool/hyla/config and
		/var/spool/hyla/config.modem 
	    until one can send and receive faxes
	(4) Further customization can be done by editing
		/usr/lib/hyla/typerules
		/usr/lib/hyla/faxcover.ps
		/var/spool/hyla/bin/faxrcvd


Let me answer some of the issues raised:

(1) /usr vs /usr/local

	Traditionally unix systems put all the programs that were
shipped with the machine /usr/bin. The programs that were installed
by the sys admin went into /usr/local. Undoubtedly, this a very useful
convention.

	However, RPMs provide a new way of managing applications. /usr
directory is no longer monolithic. It is composed of files from
the RPM packages. The philosophy of Redhat has been to put the RPM
files in /usr. /usr/local should be used for the .tar.gz packages. 
Not all packagers follow Redhat convention. Redhat itself
follows the convention meticulously and I hope Redhat will include
the hylafax RPM in their next release.


(2) fax vs hyla

	I am sorry that I had to deviate from the convention
and changed /var/spool/fax to /var/spool/hyla and /usr/lib/fax
to /usr/lib/hyla. There are two other packages that are bundled with
Redhat Linux: efax and mgetty+sendfax. mgetty+sendfax has two 
directories, /var/spool/fax/incoming and /var/spool/fax/outgoing.
I think it will be a major source of confusion to have 
6 directories of hylafax and 2 directories of mgetty+sendfax
in the same /var/spool/fax. While removing all the other
fax packages is an option, most people don't bother to
remove them. To the extent possible, I think it is better to
not to intrude into other's namespace.

(3) xferstats and xferlog

After a suggestion by Nico Garcia I changed it to the more descriptive
xferfaxstats intstead of the confusing xferstat. Similarly I changed 
xferlog to xferfaxlog. 
	

(4) default etc/config etc/config.modem

While one can use 'faxsetup' and 'faxaddmodem' to generate these
files, I found it easier to edit these files directly. Especially
'faxaddmodem' can be a source of real frustration. It should be
remembered that config and config.modem most likely need some editing
to suit the particular modem at hand.

Ramana

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Comments: Authenticated sender is <spine001@gold.tc.umn.edu>
From: Julio C Spinelli <spine001@gold.tc.umn.edu>
To: Steve Williams <steve@genie96.com>
Date: Tue, 21 Oct 1997 22:41:55 +0000
Subject: Re: flexfax: Hylafax rpm issues
CC: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

I fully agree with all your points, as I was seeing the emails on the 
RPM I was thinking of the nightmare that it was going to create on 
the users. If they have gotten confused for the warning messages on 
egetty...... That is only a hook...

I didn't intervene in the discussion because I thought that it was an 
effort sponsored (at least spiritually by Sam).

Please feel free to post this if you see it fit.

On another token, who is working on the patch? Why? I am having a 
problem while using egetty. After returning and after hylafax starts 
to answer in fax mode, I get a write error when hylafax attempts to 
write to the modem and then it apparently quits answering in fax mode 
and goes to reset the modem and wait for call again. In my real time 
experience, I never try to deal with hardware like modems only once, 
I would suggest that we should attempt to do Sys::write(fd,.....  , 
at least 10 times with 1 ms intervals if an error is reported by the 
system. In my case it is reporting errno 5 (busy resource). Similarly 
when the modem reports a response that is not the expected one, 
hylafax could be made more robust if the unsuccessful command is sent 
at least five times. Don't ask me why, but I've seen USR modems 
require this at times to respond to a command, the instances that I 
saw were during voice command mode.

By the way, I am using hylafax 4.0pl1, and before I start changing 
the source I would like to know if this has been fixed or if it will 
be fixed in the patch.

Thanks in advance for your consideration of my questions!

Best regards,

Julio C. Spinelli
Professor EE dpt. UMN

> From:          Steve Williams <steve@genie96.com>
> Subject:       Re: flexfax: Hylafax rpm issues
> To:            juvvadi@lekha.org (Ramana Juvvadi)
> Date:          Tue, 21 Oct 1997 16:52:35 -0600 (MDT)
> Cc:            flexfax@sgi.com

> Hi,
> 
> First, let me say it is nice that someone is willing to spend the time to
> put together a "rpm" for Linux.  
> 
> My second question would be..., are you willing to support it yourself
> for the next eternity?
> 
> I have been involved with HylaFAX since the old FlexFAX days.  It has been
> wonderful that for many of the issues involved, it didn't matter what
> platform the software was installed on, because all platforms were basically
> alike.  I have been the Beta Binary person for AIX for ages.
> 
> I knew where I could recommend someone to look if there was a problem, 
> regardless of the operating system.
> 
> You are reducing the ability of the mailing list to support Linux users of
> HylaFAX by arbitrairly deciding to do things that are specific to your 
> implementation.
> 
> It might come down to the point of "I don't know, contact Ramana, the guy
> who set up the RPM", because I have no idea where the "hfaxd.conf" file 
> is installed when they use your RPM.
> 
> One MAJOR issue I have is that you are recommending that people NOT RUN 
> FAXSETUP & FAXADDMODEM ???
> 
> WHAT???
> 
> These programs do EXTENSIVE testing of the system to ensure it is set up
> properly.  They probe the modem & make most of the decisions for the end user.
> 
> IF THEY CAN'T PROBE THE MODEM, THEN THERE SOMETHING WRONG WITH THE SYSTEM!!!
> AND NOTHING WILL WORK PERIOD!!!.
> 
> As to editing modem scripts by hand, that is absolutely insane.  If people
> accept the defaults to faxaddmodem, they will get a system that works.  If
> the start playing with the files themselves, god knows what they are going
> to change in a "desperate" attempt to get the modem working.
> 
> Granted, for many NO-NAME fax modems out there, the config files may have to 
> be edited, but that is not what HylaFAX was designed for.  It is a ROBUST
> fax package, that works with GOOD QUALITY fax modems.   I have NEVER had to
> manually edit a modem config file unless I made a typo in faxaddmodem that
> I didn't catch.  If you have had to manually edit a modem config file for 
> anything other than the tagline or whatever, then I wonder why?
> 
> If you are looking for a simple fax package, people should be using some other
> package.
> 
> Remember, a software package is only as good as the support that it gets,
> and these (HylaFAX )non-standard locations and naming conventions would 
> ( in my opinion ) reduce the ability of the mailing list to support the 
> package.
> 
> Also, were you aware that we are currently working on a new patch level 
> for HylaFAX?
> 
> -- 
> 	Steve Williams, Calgary, Alberta, Canada
> 	Genie Computer Systems Inc.
> 	steve@genie96.com
> 
> "A man doesn't begin to attain wisdom until he recognizes that he is 
>  no longer indespensable."
> - Admiral Richard E. Byrd ( 1888-1957 )
> ----------------------------------------------------------------------
> > 
> > I hope with this rpm release, hylafax is easier than before to setup.
> > 
> > The old annoyances of afm directories, entries in /etc/aliases, 
> > /etc/inetd.conf, /etc/inittab, /etc/services, creating of the user
> > fax are all taken care of.
> > 
> > To send and receive faxes all one needs to do is
> > 
> > 	(1) rpm -i hylafax-4.0-4.i386.rpm
> > 	(2) reboot
> > 	(3) Edit /var/spool/hyla/config and
> > 		/var/spool/hyla/config.modem 
> > 	    until one can send and receive faxes
> > 	(4) Further customization can be done by editing
> > 		/usr/lib/hyla/typerules
> > 		/usr/lib/hyla/faxcover.ps
> > 		/var/spool/hyla/bin/faxrcvd
> > 
> > 
> > Let me answer some of the issues raised:
> > 
> > (1) /usr vs /usr/local
> > 
> > 	Traditionally unix systems put all the programs that were
> > shipped with the machine /usr/bin. The programs that were installed
> > by the sys admin went into /usr/local. Undoubtedly, this a very useful
> > convention.
> > 
> > 	However, RPMs provide a new way of managing applications. /usr
> > directory is no longer monolithic. It is composed of files from
> > the RPM packages. The philosophy of Redhat has been to put the RPM
> > files in /usr. /usr/local should be used for the .tar.gz packages. 
> > Not all packagers follow Redhat convention. Redhat itself
> > follows the convention meticulously and I hope Redhat will include
> > the hylafax RPM in their next release.
> > 
> > 
> > (2) fax vs hyla
> > 
> > 	I am sorry that I had to deviate from the convention
> > and changed /var/spool/fax to /var/spool/hyla and /usr/lib/fax
> > to /usr/lib/hyla. There are two other packages that are bundled with
> > Redhat Linux: efax and mgetty+sendfax. mgetty+sendfax has two 
> > directories, /var/spool/fax/incoming and /var/spool/fax/outgoing.
> > I think it will be a major source of confusion to have 
> > 6 directories of hylafax and 2 directories of mgetty+sendfax
> > in the same /var/spool/fax. While removing all the other
> > fax packages is an option, most people don't bother to
> > remove them. To the extent possible, I think it is better to
> > not to intrude into other's namespace.
> > 
> > (3) xferstats and xferlog
> > 
> > After a suggestion by Nico Garcia I changed it to the more descriptive
> > xferfaxstats intstead of the confusing xferstat. Similarly I changed 
> > xferlog to xferfaxlog. 
> > 	
> > 
> > (4) default etc/config etc/config.modem
> > 
> > While one can use 'faxsetup' and 'faxaddmodem' to generate these
> > files, I found it easier to edit these files directly. Especially
> > 'faxaddmodem' can be a source of real frustration. It should be
> > remembered that config and config.modem most likely need some editing
> > to suit the particular modem at hand.
> > 
> > Ramana
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Steve Williams <steve@genie96.com>
Subject: Re: flexfax: Hylafax rpm issues
To: juvvadi@lekha.org (Ramana Juvvadi)
Date: Tue, 21 Oct 1997 16:52:35 -0600 (MDT)
Cc: flexfax@sgi.com
Read-Receipt-To: steve@genie96.com
Sender: owner-flexfax@celestial.com

Hi,

First, let me say it is nice that someone is willing to spend the time to
put together a "rpm" for Linux.  

My second question would be..., are you willing to support it yourself
for the next eternity?

I have been involved with HylaFAX since the old FlexFAX days.  It has been
wonderful that for many of the issues involved, it didn't matter what
platform the software was installed on, because all platforms were basically
alike.  I have been the Beta Binary person for AIX for ages.

I knew where I could recommend someone to look if there was a problem, 
regardless of the operating system.

You are reducing the ability of the mailing list to support Linux users of
HylaFAX by arbitrairly deciding to do things that are specific to your 
implementation.

It might come down to the point of "I don't know, contact Ramana, the guy
who set up the RPM", because I have no idea where the "hfaxd.conf" file 
is installed when they use your RPM.

One MAJOR issue I have is that you are recommending that people NOT RUN 
FAXSETUP & FAXADDMODEM ???

WHAT???

These programs do EXTENSIVE testing of the system to ensure it is set up
properly.  They probe the modem & make most of the decisions for the end user.

IF THEY CAN'T PROBE THE MODEM, THEN THERE SOMETHING WRONG WITH THE SYSTEM!!!
AND NOTHING WILL WORK PERIOD!!!.

As to editing modem scripts by hand, that is absolutely insane.  If people
accept the defaults to faxaddmodem, they will get a system that works.  If
the start playing with the files themselves, god knows what they are going
to change in a "desperate" attempt to get the modem working.

Granted, for many NO-NAME fax modems out there, the config files may have to 
be edited, but that is not what HylaFAX was designed for.  It is a ROBUST
fax package, that works with GOOD QUALITY fax modems.   I have NEVER had to
manually edit a modem config file unless I made a typo in faxaddmodem that
I didn't catch.  If you have had to manually edit a modem config file for 
anything other than the tagline or whatever, then I wonder why?

If you are looking for a simple fax package, people should be using some other
package.

Remember, a software package is only as good as the support that it gets,
and these (HylaFAX )non-standard locations and naming conventions would 
( in my opinion ) reduce the ability of the mailing list to support the 
package.

Also, were you aware that we are currently working on a new patch level 
for HylaFAX?

-- 
	Steve Williams, Calgary, Alberta, Canada
	Genie Computer Systems Inc.
	steve@genie96.com

"A man doesn't begin to attain wisdom until he recognizes that he is 
 no longer indespensable."
- Admiral Richard E. Byrd ( 1888-1957 )
----------------------------------------------------------------------
> 
> I hope with this rpm release, hylafax is easier than before to setup.
> 
> The old annoyances of afm directories, entries in /etc/aliases, 
> /etc/inetd.conf, /etc/inittab, /etc/services, creating of the user
> fax are all taken care of.
> 
> To send and receive faxes all one needs to do is
> 
> 	(1) rpm -i hylafax-4.0-4.i386.rpm
> 	(2) reboot
> 	(3) Edit /var/spool/hyla/config and
> 		/var/spool/hyla/config.modem 
> 	    until one can send and receive faxes
> 	(4) Further customization can be done by editing
> 		/usr/lib/hyla/typerules
> 		/usr/lib/hyla/faxcover.ps
> 		/var/spool/hyla/bin/faxrcvd
> 
> 
> Let me answer some of the issues raised:
> 
> (1) /usr vs /usr/local
> 
> 	Traditionally unix systems put all the programs that were
> shipped with the machine /usr/bin. The programs that were installed
> by the sys admin went into /usr/local. Undoubtedly, this a very useful
> convention.
> 
> 	However, RPMs provide a new way of managing applications. /usr
> directory is no longer monolithic. It is composed of files from
> the RPM packages. The philosophy of Redhat has been to put the RPM
> files in /usr. /usr/local should be used for the .tar.gz packages. 
> Not all packagers follow Redhat convention. Redhat itself
> follows the convention meticulously and I hope Redhat will include
> the hylafax RPM in their next release.
> 
> 
> (2) fax vs hyla
> 
> 	I am sorry that I had to deviate from the convention
> and changed /var/spool/fax to /var/spool/hyla and /usr/lib/fax
> to /usr/lib/hyla. There are two other packages that are bundled with
> Redhat Linux: efax and mgetty+sendfax. mgetty+sendfax has two 
> directories, /var/spool/fax/incoming and /var/spool/fax/outgoing.
> I think it will be a major source of confusion to have 
> 6 directories of hylafax and 2 directories of mgetty+sendfax
> in the same /var/spool/fax. While removing all the other
> fax packages is an option, most people don't bother to
> remove them. To the extent possible, I think it is better to
> not to intrude into other's namespace.
> 
> (3) xferstats and xferlog
> 
> After a suggestion by Nico Garcia I changed it to the more descriptive
> xferfaxstats intstead of the confusing xferstat. Similarly I changed 
> xferlog to xferfaxlog. 
> 	
> 
> (4) default etc/config etc/config.modem
> 
> While one can use 'faxsetup' and 'faxaddmodem' to generate these
> files, I found it easier to edit these files directly. Especially
> 'faxaddmodem' can be a source of real frustration. It should be
> remembered that config and config.modem most likely need some editing
> to suit the particular modem at hand.
> 
> Ramana

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 00:27:04 -0400 (EDT)
From: Dave Wreski <dave@nic.com>
To: Steve Williams <steve@genie96.com>
Subject: Re: flexfax: Hylafax rpm issues
Cc: flexfax@sgi.com, (Ramana Juvvadi) <juvvadi@lekha.org>
Sender: owner-flexfax@celestial.com


> First, let me say it is nice that someone is willing to spend the time to
> put together a "rpm" for Linux.  

Hmm.. From the rest of your message, that sounds like a loaded compliment.

> My second question would be..., are you willing to support it yourself
> for the next eternity?

Are you actually a linux user?  Everyone knows the package maintainer is simply
the package maintainer, not the content.

> I have been involved with HylaFAX since the old FlexFAX days.  It has been
> wonderful that for many of the issues involved, it didn't matter what
> platform the software was installed on, because all platforms were basically
> alike.  I have been the Beta Binary person for AIX for ages.

So where was your help with the package when he was building it?

> I knew where I could recommend someone to look if there was a problem, 
> regardless of the operating system.
> 
> You are reducing the ability of the mailing list to support Linux users of
> HylaFAX by arbitrairly deciding to do things that are specific to your 
> implementation.

I think there was some previous conversation about the changes he was going to
make to the package before he put it together, that you must have missed
objecting to then.

> It might come down to the point of "I don't know, contact Ramana, the guy
> who set up the RPM", because I have no idea where the "hfaxd.conf" file 
> is installed when they use your RPM.

I didn't when attempting to use the last RPM that was produced.

> One MAJOR issue I have is that you are recommending that people NOT RUN 
> FAXSETUP & FAXADDMODEM ???
> 
> These programs do EXTENSIVE testing of the system to ensure it is set up
> properly.  They probe the modem & make most of the decisions for the end
> user.

Wouldn't it be nice to attempt to make hylafax as easy to use and
install/maintain as possible?  After all, he listed the implications of using
the package and what changes were made.  This is not a substitute for not
reading the documentation.

> IF THEY CAN'T PROBE THE MODEM, THEN THERE SOMETHING WRONG WITH THE SYSTEM!!!
> AND NOTHING WILL WORK PERIOD!!!.

OH MY GOSH-GOLLY!  Somehow I bet he tested it on his box first, and it most
likely was one step closer to being perfect than the last RPM that was built. 
Again, he can suggest that you don't probe the modem, but that doesn't mean you
don't have to read the documentation, and probe the modem yourself.

> As to editing modem scripts by hand, that is absolutely insane.  If people
> accept the defaults to faxaddmodem, they will get a system that works.  If
> the start playing with the files themselves, god knows what they are going
> to change in a "desperate" attempt to get the modem working.

And the earth is round.  Ha, no one would ever think something could be made
easier to maintain.  Apparently RPM == Linux idiot.

> Granted, for many NO-NAME fax modems out there, the config files may have to 
> be edited, but that is not what HylaFAX was designed for.  It is a ROBUST
> fax package, that works with GOOD QUALITY fax modems.   I have NEVER had to
> manually edit a modem config file unless I made a typo in faxaddmodem that
> I didn't catch.  If you have had to manually edit a modem config file for 
> anything other than the tagline or whatever, then I wonder why?

And don't you think the HylaFAX maintainers would be happy to see someone else
got his piece of shit modem working with it?  There are no stipluations as to
which modems absolutely must not be used under no circumstances, because there
is a chance you may have to edit a text file or two.

> If you are looking for a simple fax package, people should be using some
> other
> package.

Yeah, like Windows FAX, right?  After all, this is meant to be difficult, and
only Windows programs are easy to use.

> Remember, a software package is only as good as the support that it gets,
> and these (HylaFAX )non-standard locations and naming conventions would 
> ( in my opinion ) reduce the ability of the mailing list to support the 
> package.

Again, who says support is bound to the package maintainer?  I'm sure he's
included the documentation with it.

> Also, were you aware that we are currently working on a new patch level 
> for HylaFAX?

Possibly you should have spoken up last week when he announced development of
the new RPM.  Besides, are you discouraging him from getting a head start on
development?  Do you know exactly when its going to be released?  Possibly it
might help a few people in the meantime.

Again, the reason for building an RPM is to simplify the build process, and
make an attempt at making configuration easier.  There is nothing to bound him
to this package, and it certainly isn't a substitute for reading the manual.

After all, do we all go to RedHat after they bundle samba, and it doesn't work
out-of-the-box with our servers?  Oh my gosh, I have to edit the samba
configuration file, and configure my server name.  Who's the President of
RedHat Software, Inc?  I want his direct phone number!  I want my money back.

Dave

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Steve Williams <steve@genie96.com>
Subject: Re: flexfax: Hylafax rpm issues
To: dave@nic.com (Dave Wreski)
Date: Wed, 22 Oct 1997 01:59:43 -0600 (MDT)
Cc: steve@genie96.com, flexfax@sgi.com, juvvadi@lekha.org
Read-Receipt-To: steve@genie96.com
Sender: owner-flexfax@celestial.com

Howdy,

[snip]

> 
> Are you actually a linux user? Everyone knows the package maintainer is simply
> the package maintainer, not the content.

I have dabbled with Linux.  We are an IBM house here, so that is where my
expertise lies, AIX & SCO.  Regardless, the person that "creates" the package 
dictates where files are going to go.  They will ( hopefully ) also include
installation notes that NO-ONE EVER READS... RTFM <g>.

[snip]

> I think there was some previous conversation about the changes he was going to
> make to the package before he put it together, that you must have missed
> objecting to then.

No, I didn't miss it.  At the time, I didn't think it was that big of an
issue.  When it is mentioned that it was hoped that this RPM would be 
included in the RedHat distribution of Linux, then that places a much 
greater importance on the issue than before ( in my mind, anyway ).

Anyone that has to pick up the package of the Web or an FTP site at least 
has a bit of knowlege of their system, because it is already connected to 
the Internet ( well, that is a LARGE assumption, but not TOO outrageous ).

When you give a nice "simple" CD-ROM to someone to install on their PC, and
I must say, the Linux installation blows the door off of SCO, then it
is not quite so obvious that they are super knowlegable about their system.
Granted, it doesn't preclude it either.

[snip]

> > One MAJOR issue I have is that you are recommending that people NOT RUN 
> > FAXSETUP & FAXADDMODEM ???
> > 
> > These programs do EXTENSIVE testing of the system to ensure it is set up
> > properly.  They probe the modem & make most of the decisions for the end
> > user.
> 
> Wouldn't it be nice to attempt to make hylafax as easy to use and
> install/maintain as possible?  After all, he listed the implications of using
> the package and what changes were made.  This is not a substitute for not
> reading the documentation.
> 

Give me a break!!!.  Unfortunatly, I deleted the message, but just today,
some email came across the mailing list that someone thought this RPM would
"DO EVERYTHING FOR HIM".

> > IF THEY CAN'T PROBE THE MODEM, THEN THERE SOMETHING WRONG WITH THE SYSTEM!!!
> > AND NOTHING WILL WORK PERIOD!!!.
> 
> OH MY GOSH-GOLLY!  Somehow I bet he tested it on his box first, and it most
> likely was one step closer to being perfect than the last RPM that was built. 
> Again, he can suggest that you don't probe the modem, but that doesn't mean you
> don't have to read the documentation, and probe the modem yourself.
> 

Get real!!!.  People want to read as little as possible ( including myself ).
If someone says to do it this way, they will take that as the gosphel truth,
and do it that way.  The fact that there is AMPLE other documentation will
not affect their decision.  I am just as bad as everyone else.  I try to 
do something first, then RTFM.  If the RTFM (with the RPM ) says to manually 
edit the files, that's what they will do.  If they can't figure it out 
after that, unless there is a big pointer in the README with the RPM, then
off to the mailing list they will go.

faxsetup and faxaddmodem work!!  If they don't work, then there is something
fundamentally wrong with your system that will jump out and bite you sometime
down the road.  If the shell script breaks ( earlier versions of bash ), 
then there is a good chance that some other shell script is going to break,
as HylaFAX relies HEAVILY on shell scripting.  So you get the modem working
by manually editing the config files.  Then people can't figure out why they
aren't getting email notification of their faxes when the "notify.sh" script
won't run ( just using this as an example ).

If the faxaddmodem script hangs during running of it, then there is a very good 
chance that something is wrong with how the modem is configured, __OR__ at
the operating system level.  ( Early versions of AIX had problems at the OS
level with some things that caused faxaddmodem to hang..., AND other programs
as well ).

[snip]

> 
> And the earth is round.  Ha, no one would ever think something could be made
> easier to maintain.  Apparently RPM == Linux idiot.
> 

Ummm, no I am not saying Linux idiot, but not necessarily an indepth 
knowlegable user.  I only say this because IT HAS SUCH A GREAT INSTALL
PACKAGE!!.  Anyone can install pretty much anything easily.

The days of having to hack code from the internet to get it to compile on
another platform are pretty much over.  ( I just compiled the sh-utils-1.16
on AIX 3.2.5 with no errors ).  Things are at the point where you no longer
need to be a programmer to be grabbing software off of the internet, just
follow the directions.  This means that the user may not have a fundamental
understanding of *nix operating systems, as MAYBE they are coming from
the Windows environment, and using Linux as a learning environment.

I am all for getting some things easier to install, something as simple as
configuring the syslogd to produce debugging messages took me ages to figure
out when I first tackled the job.  At that time, there were no binary 
distributions, or README's.  But, I do understand the syslog.conf file 
now<g>.

That agony should NOT be inflicted on a "novice" user.  BUT!!  They should 
still know something about it, because what if the see "daemon" in their
config files that they are editing by hand.  If they don't know how the
debugging works, they might change that to some other item, and voilla, 
their debugging quits working.  If they have no idea that they RPM "tweaked"
their syslog.conf file, they will have no idea what broke where.

[snip]

> 
> And don't you think the HylaFAX maintainers would be happy to see someone else
> got his piece of shit modem working with it?  There are no stipluations as to
> which modems absolutely must not be used under no circumstances, because there
> is a chance you may have to edit a text file or two.
> 

The people that can get a "piece of *hit" modem working with HylaFAX are not
likely to be the ones that are going to be needing ELEMENTARY help.  If 
someone is comfortable with playing with MSB2LSB & all those wonderful things 
that I have no idea about, then they are more than likely to read manuals/web 
pages for assistance before jumping on the mailing list.

I am concerned about the users that install HylaFAX on their home PC with
no fundamental understanding of what is going on because the RPM did 
everything for them, and then they hit the HylaFAX mailing list for help.

If they haven't even run faxsetup or faxaddmodem ( which kinda give clues
about what is wrong ), how can we help them on the mailing list?

> > If you are looking for a simple fax package, people should be using some
> > other package.
> 
> Yeah, like Windows FAX, right?  After all, this is meant to be difficult, and
> only Windows programs are easy to use.
> 

Nope, I was referring to mgetty+sendfax.  Windows is a REAL *HITTY 
environment period!!  But, mgetty+sendfax doesn't support Class 1 Fax, 
which HylaFAX does.  Thus, I would expect that would motivate even more
people to use HylaFAX vs. mgetty+sendfax.  It seems that most modem chip 
manufacturers have dumped Class 2 fax when they had to fit the 56K onto 
the ROM, ( and even some of the 33.6's ) so most of the new "bottom end" 
modems are all Class 1.

> > Remember, a software package is only as good as the support that it gets,
> > and these (HylaFAX )non-standard locations and naming conventions would 
> > ( in my opinion ) reduce the ability of the mailing list to support the 
> > package.
> 
> Again, who says support is bound to the package maintainer?  I'm sure he's
> included the documentation with it.

Yes, but there are lots of us on the mailing list that will never read the
Linux documentation.  And the mailing list is where you will get resources
from people that have been using the software for years.  Ask me what 
ModemClass is used for!!  It is DOES NOT indicate Class 1, 2, 2.0.  If 
someone get's that one wrong, good luck sending a fax!!!  I have seen this
one come across the mailing list many a time, & I have been able to 
help them out because I debugged the code on one of the beta versions where
there was a problem.

If there are too many "RTMF" questions on the mailing list, I would just 
unsubscribe from the mailing list & stay on the "beta tester" mailing list.
That would be a shame, because I like to return the favors that I have had 
bestowed on me when I have required assistance.  However, one can get 
tired of silly questions. 

[snip]
> 
> Again, the reason for building an RPM is to simplify the build process, and
> make an attempt at making configuration easier.  There is nothing to bound him
> to this package, and it certainly isn't a substitute for reading the manual.
> 
Simplifying the build process is grand.  I'm all for it.  My problem I guess
is that DON'T SKIP STEPS.  If there is a readme with the RPM, that is going
to be what people read, simply because there is so much HylaFAX documentation
one does not know where to start.  Suggesting to edit config files by hand
rather than running the supplied utilities is supposed to be easier??  What
about testing to see if the cabling is correct on the modem ( which 
faxaddmodem kinda does )?  

As for reading the manual...

I installed Linux on a 486 without reading a single manual.  A guy gave me 
the floppies & a CD & said go to it.  It's an amazing process.  I don't
really need to know much about hardware to get it going.  Those are the
people that I worry about.  They have a el-cheapo Class 1 fax modem, so 
they install the HylaFAX RPM, and can't get it working, & flood the mailing
list.  The "hackers" that play around with the kernel aren't so much a 
concern.
 
[oops, Sorry, I deleted the reference to installing samba not working]

Fax Modems are MUCH more fickle than samba.  You have byte order, flow 
control, modems that switch baud rate to 19200 when the recieve a fax,
and god knows how many other "features".  A novice user is going to have
many more problems than just editing a samba config file & changing
a server name.

Anyway, I have spewed on enough, I have to go home & go to bed.

It's been nice chatting.
-- 
	Steve Williams, Calgary, Alberta, Canada
	Genie Computer Systems Inc.
	steve@genie96.com

"A man doesn't begin to attain wisdom until he recognizes that he is 
 no longer indespensable."
- Admiral Richard E. Byrd ( 1888-1957 )

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
To: Dave Wreski <dave@nic.com>
cc: Steve Williams <steve@genie96.com>, flexfax@sgi.com,
        juvvadi@lekha.org (Ramana Juvvadi)
Subject: Re: flexfax: Hylafax rpm issues 
Date: Wed, 22 Oct 1997 10:46:26 +0100
From: "Mr. Arlington Hewes" <tpcadmin@info.tpc.int>
Sender: owner-flexfax@celestial.com

>>>>> On Wed, 22 Oct 1997, "DW" == Dave Wreski wrote:

  +> My second question would be..., are you willing to support it yourself
  +> for the next eternity?

  DW> Are you actually a linux user?  Everyone knows the package maintainer
  DW> is simply the package maintainer, not the content.

I must agree with Steve here, Ramana has set the precedent of deviating from
traditional HylaFAX build conventions, even to the point of renaming one
of the scripts. In effect, he has become the content maintainer. There's
no telling where this will lead. As a RedHat Linux user (before you start
trying to drag me through the mud), I'm concerned that the 'official' RPM
will be different enough from the rest of the packages that it will not be
possible for us to support newbies who speak up in this forum.

Changes such as renaming xferstats should be addressed in this forum, and 
changed in the master distribution, if justified.

I support much of what Ramana has done. In particular, sticking to RPM 
tradition and wishing to be included in the next RedHat release are 
admirable ambitions, and will do much to further the cause of us HylaFAX 
evangelists. However, I do feel he's being a bit reactionary in making 
sweeping changes based on minimal discussion.

-DPN

-- 
Mr. Arlington Hewes           (tpcadmin@info.tpc.int)
The TPC.INT Subdomain          (http://www.tpc.int/)

       **************************************************
       ***  FOR GENERAL INFORMATION                   ***
       ***         Mail to tpcfaq@info.tpc.int        ***
       ***  FOR A LIST OF CURRENT COVERAGE            ***
       ***         Mail to tpccover@info.tpc.int      ***
       **************************************************

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Hylafax rpm issues
To: spine001@gold.tc.umn.edu (Julio C Spinelli)
Date: Wed, 22 Oct 1997 16:18:52 +0200 (MSZ)
Cc: steve@genie96.com, flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Julio C Spinelli wrote:
    
    I fully agree with all your points, as I was seeing the emails on the 
    RPM I was thinking of the nightmare that it was going to create on 
    the users. If they have gotten confused for the warning messages on 
    egetty...... That is only a hook...
    
    I didn't intervene in the discussion because I thought that it was an 
    effort sponsored (at least spiritually by Sam).
    
    Please feel free to post this if you see it fit.
    
    On another token, who is working on the patch? Why? I am having a 

I'm working on the new patch-level of HylaFAX (it's already
running as a beta on my system at home). It will contain
a lot of the fixes sent to this list or to me directly (and come
as usual with a HTML page describing all changes). This beta
will go through a beta-test phase in the furture (there are
some people who are willing to do this and prepare their
binary distributions).

    problem while using egetty. After returning and after hylafax starts 
    to answer in fax mode, I get a write error when hylafax attempts to 
    write to the modem and then it apparently quits answering in fax mode 
    and goes to reset the modem and wait for call again. In my real time 
    experience, I never try to deal with hardware like modems only once, 
    I would suggest that we should attempt to do Sys::write(fd,.....  , 
    at least 10 times with 1 ms intervals if an error is reported by the 
    system. In my case it is reporting errno 5 (busy resource). Similarly 
    when the modem reports a response that is not the expected one, 
    hylafax could be made more robust if the unsuccessful command is sent 
    at least five times. Don't ask me why, but I've seen USR modems 
    require this at times to respond to a command, the instances that I 
    saw were during voice command mode.
    
    By the way, I am using hylafax 4.0pl1, and before I start changing 
    the source I would like to know if this has been fixed or if it will 
    be fixed in the patch.

I got your question about this egetty-problem and will see what I can
do. No promice (and also no solution if I can't reproduce it on my
system). If your OS has a tool like truss (SVR4) or strace (Linux)
try digging into it by trussing/stracing the faxgetty(1M) proc or
send such a file to me.

	matthias
    
    Thanks in advance for your consideration of my questions!
    
    Best regards,
    
    Julio C. Spinelli
    Professor EE dpt. UMN

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 10:21:50 -0400
From: Ramana Juvvadi <juvvadi@lekha.org>
To: "Mr. Arlington Hewes" <tpcadmin@info.tpc.int>
Cc: Dave Wreski <dave@nic.com>, Steve Williams <steve@genie96.com>,
        flexfax@sgi.com, juvvadi@lekha.org (Ramana Juvvadi)
Subject: Re: flexfax: Hylafax rpm issues 
Sender: owner-flexfax@celestial.com

Mr. Arlington Hewes writes:


 > tradition and wishing to be included in the next RedHat release are 
 > admirable ambitions, and will do much to further the cause of us HylaFAX 
 > evangelists. However, I do feel he's being a bit reactionary in making 
 > sweeping changes based on minimal discussion.

The only "content" change I made is xferstat --> xferfaxstat
xferlog --> xferfaxlog because it is clashing with wu-ftpd. 
If it is addressed in the next version or if somebody
has a better suggestion I am willing to change it.  

Ramana

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 11:52:58 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
Cc: Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: Hylafax rpm issues
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Tue, 21 Oct 1997, Steve Williams wrote:

> I knew where I could recommend someone to look if there was a problem, 
> regardless of the operating system.

Yeah, always useful.

> You are reducing the ability of the mailing list to support Linux users of
> HylaFAX by arbitrairly deciding to do things that are specific to your 
> implementation.

Umm: This is unfair. It's volunteer work, and a significant chunk of it.
As long as the quality is reasonable, that should work.

> It might come down to the point of "I don't know, contact Ramana, the guy
> who set up the RPM", because I have no idea where the "hfaxd.conf" file 
> is installed when they use your RPM.

What's wrong with "Read The Fine Manual"? 

> One MAJOR issue I have is that you are recommending that people NOT RUN 
> FAXSETUP & FAXADDMODEM ???
> 
> WHAT???
> 
> These programs do EXTENSIVE testing of the system to ensure it is set up
> properly.  They probe the modem & make most of the decisions for the end user.

Good points! Especially given the need for PnP setup on many modem lines,
running setserial, etc. under Linux systems, these are basic utilities to
test modem function.

> As to editing modem scripts by hand, that is absolutely insane.  If people
> accept the defaults to faxaddmodem, they will get a system that works.  If
> the start playing with the files themselves, god knows what they are going
> to change in a "desperate" attempt to get the modem working.

May I disagree strongly? Until and unless the modem configuration
files are correct for the particular version of the modem, etc., the
config files need to be corrected at the outset. For example: many
Rockwell chipset 14.4 modems have problems with zero padding without
certain options set. It is *vastly* easier to change the default
configuration file once than to change it again and again with
faxaddmodem anytime you set up a new modem or re-initialize it.

> Granted, for many NO-NAME fax modems out there, the config files may have to 
> be edited, but that is not what HylaFAX was designed for.  It is a ROBUST
> fax package, that works with GOOD QUALITY fax modems.   I have NEVER had to

Freeware that demands top of the line modems is not freeware.

> manually edit a modem config file unless I made a typo in faxaddmodem that
> I didn't catch.  If you have had to manually edit a modem config file for 
> anything other than the tagline or whatever, then I wonder why?

Because it's a pain in the neck to run faxaddmodem and re-enter all
the weird fields for your local setup.

> Remember, a software package is only as good as the support that it gets,
> and these (HylaFAX )non-standard locations and naming conventions would 
> ( in my opinion ) reduce the ability of the mailing list to support the 
> package.

Also good points. Using "/usr/hyla" is silly. Use "/usr/hylafax", etc.

Under SunOS, unfortunately, the /var/spool/hylafax directory is too
small to support the fax spool. So I carefully re-arranged it to go in
the /usr/local/spool/hylafax directory, and carefully published the
modifications to the config file in the SunOS installation notes. So
it is sometimes necessary....

This can be a problem. I had to modify faxtool for local
configurations to get it to find the faxes correctly. As long as the
system-based modifications are done consistently and with good annotation,
I don't see the problem.


> > (4) default etc/config etc/config.modem
> > 
> > While one can use 'faxsetup' and 'faxaddmodem' to generate these
> > files, I found it easier to edit these files directly. Especially
> > 'faxaddmodem' can be a source of real frustration. It should be
> > remembered that config and config.modem most likely need some editing
> > to suit the particular modem at hand.

Yeah, especially since things like the received file permissions and
local area codes are stored in *four* files that overwrite each other:
config/skel, config/class1, config/class2, and config/class2.0. This
can be *very* confusing.

I've submitted patches for this little confusion....

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNE4hVD/+ItycgIJRAQEVTwP/dTlEz+B0IjoBTIi0BG7pjaw1NEQc8htR
0rbJVaiXNdCmYFaq/DZ79nITk2NCkKF4zFT9TiiE/LXH3S/Y+xbDyElK/F10f5Ms
iTk2HnIM7X7pQJtagTKQ7Ko28p3+VrafUhsAPf9GxITJpzb+JxXAmWqrwclqbq6H
hBU57sSaLvk=
=SLsH
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 12:01:35 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: "Mr. Arlington Hewes" <tpcadmin@info.tpc.int>
Cc: Nico Garcia <flexfax@sgi.com>, Ramana Juvvadi <juvvadi@lekha.org>
Subject: Re: flexfax: Hylafax rpm issues 
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Wed, 22 Oct 1997, Mr. Arlington Hewes wrote:

> I must agree with Steve here, Ramana has set the precedent of deviating from
> traditional HylaFAX build conventions, even to the point of renaming one
> of the scripts. In effect, he has become the content maintainer. There's
> no telling where this will lead. As a RedHat Linux user (before you start
> trying to drag me through the mud), I'm concerned that the 'official' RPM
> will be different enough from the rest of the packages that it will not be
> possible for us to support newbies who speak up in this forum.
> 
> Changes such as renaming xferstats should be addressed in this forum, and 
> changed in the master distribution, if justified.

First, it's not "official" yet. Second, all against renaming
"xferstats" to prevent overwriting Linux programs, please raise their
mouse pads?

Let me see how busy I am this week: maybe I can be a good test bed for
the Linux with my home box....

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNE4jUz/+ItycgIJRAQHtpQQAj/1c1OIOCLra+21MqI9ZUyBqR3j0mD1h
IUkfS4/dGUu8JKXcC+oY+M/UsFu42t0jkfVg+J+03lNBXmSjT3K3KAU9hQMjj5o7
gYy7CFVo/WOHsVUQd0XFtBdSKT44vtjT4yp/l3GJVYP5wQHemI9BYgzlldZfaNK7
7PUQtRSoRU8=
=muze
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 12:51:40 -0400
From: Ramana Juvvadi <juvvadi@lekha.org>
To: "Mr. Arlington Hewes" <tpcadmin@info.tpc.int>
Cc: Dave Wreski <dave@nic.com>, Steve Williams <steve@genie96.com>,
        flexfax@sgi.com, juvvadi@lekha.org (Ramana Juvvadi)
Subject: Re: flexfax: Hylafax rpm issues 
Sender: owner-flexfax@celestial.com



Disclaimer:
----------
I am only trying to build the RPM just the way Redhat does, so
that Redhat would consider including it in their distribution.
So far, I haven't contacted anybody from Redhat.  


I am surprised by the amount of controversy my rpm created. Perhaps 
I should emphasize once more my objective in building this RPM.

	(1) I want hylafax to work out of the box with minimal effort 
		for the end user. 

	(2) I want Redhat to include hylafax in their distribution. So I
		want to make it as Redhat compliant as possible.
	
I am a reasonable person and open to suggestions. There is no point 
in getting into a fight for minor points.



Let me address some of the objections that were raised about
the RPM. Please consider each issue independently and 



Skipping 'faxsetup' and 'faxaddmodem'
------------------------------------

This was probably the most controversial issue. 

I didn't have much trouble with 'faxsetup'. But rpm postinstall script
already handles faxsetup's duties. 

The recommendation that I made 'faxaddmodem' was based on my
frustration with it as an end user. I had to almost understand half of
the 'faxaddmodem' script to figure why it was hanging. I tried with 3
modems: a US Robotics Sportster, Cardinal MVPV34I, a Zoltrix modem. In
no case, could 'faxaddmodem' sniff all the parameters of the
modems. It is pointless to argue that their firmware is faulty and not
the script. Lot of potential hylafax users have these modems. 

This was very frustrating because I could send and receive with a 
simple package like efax using the same Cardinal modem. After I looked
at the config file it produced, I was wondering whether all the 
trouble I had was worth it. It would have been simpler if I had 
a template config file and tweaked one or two parameters to get
it working.

Perhaps my experience was unusual. Perhaps 'faxaddmodem' works
flawlessly for most people. On the other hand perhaps many
people gave up when it didn't work for them and never reported
in this mailing list. I know that I tried to make it work six
months ago. I just gave up and moved onto something else.

I don't understand why it is any easier to throw a question like

Getty Args	[-h %l dx_%s]?

than editing the same line in config file. I bet most people don't
have any clue about what argument getty takes. At least they know that
if the config file was produced for Linux, they may not need to change
it.

Finally, it is not as if I removed the scripts 'faxsetup' and 'faxaddmodem'. 
People can still use them to generate the config files for them. 


Change of directories and names
-------------------------------

	(a) /usr/local to /usr

		This is by Redhat's convention. We are not going to
		get Redhat to bundle any package with files in
		/usr/local. Of all the issues that were raised
		this is the only one I insist on. I am flexible
		on every other issue.

	(b) /var/spool/fax to /var/spool/hyla
		

		I don't have a strong opinon on this. I thought of
		changing it only because it was conflicting with 
		other packages like mgetty+sendfax. If there is
		a strong opinion on keeping it the way it is,
		I am willing to do so.

	(c) xferstats to xferfaxstats 

		This is prompted by moving from /usr/local to /usr
		IMO, even if it is kept in /usr/local, the name should
		be changed to a more unique name. However, I don't
		want to be perceived as "reactionary" or "arbitrary".
		I am flexible and open to suggestions.

It is important to remember that hylafax has to peacefully coexist
with other programs. It is especially important in Linux becomes
it comes with lot of software bundled on it. The less intrusive
hylafax is, the more successful will it be.

Ramana


 
	
		

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: edw@detel.com (Ed Weinberg)
To: Ramana Juvvadi <juvvadi@lekha.org>
Cc: "Mr. Arlington Hewes" <tpcadmin@info.tpc.int>, Dave Wreski <dave@nic.com>,
        Steve Williams <steve@genie96.com>, flexfax@sgi.com,
        juvvadi@lekha.org (Ramana Juvvadi)
Subject: Re: flexfax: Hylafax rpm issues 
Date: Wed, 22 Oct 1997 19:00:35 GMT
Reply-To: edw@detel.com
Sender: owner-flexfax@celestial.com

On Wed, 22 Oct 1997 12:51:40 -0400, Ramana Juvvadi <juvvadi@lekha.org>
wrote:

>	(1) I want hylafax to work out of the box with minimal effort 
>		for the end user. 

Great!  Then maybe you can answer a few more questions.

While I am not a newbe to computers, I am to hyafax, so if you are
looking for a person who does not know what he is doing to install
this package, I might be your guinie pig!  :o)

I have an RH 4.1 Linux system with a Zoom 28.8 fax modem installed.  I
know that it uses a Rockwell chipset, but am not sure which.

I installed your hylafax RPM and have upgraded it...probably twice.  I
now have hylafax-4.0-5 installed.
also:
ghostscript-5.03-2
libgr-2.0.14-2

ps tells me that the following daemon is running:
  225  ?  S    0:00 /usr/sbin/faxq

I am not sure what config files I am supposed to change.  The message
on install mentions some thiings, but I am not sure where they are
documented.  When the original modem setup program ran (before your
RPM's) , the program always hung and I got out of it with a ^C.   

Is there documentation as to what I have to do next (besides reboot)?

Currently I am try ing to use sendfax.  The telephone line attached to
the modem requires a "9" to dial out and to dial my fax number for
testing I need to dial a "1".   here is what I tried:
#sendfax -d 912033674005 /etc/hosts

Here is the reasponse:
Warning, No "hylafax" service definition, using default 4559/tcp.
Can not reach server at host "localhost", port 4559.
(end of response)

This is the farthest I have gotten!

Is my command line correct?

What else do I have to do?

Can you suggest specific docs to read?

Thank you for your help.

  --  Ed Weinberg,
      Detel, Inc., An Internet Presence Provider
      edw@detel.com
      www.detel.com/
      www.serverking.com
      www.q5.com/  <-- find someone to CoolTalk or chat with here

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 16:15:10 -0400
From: Ramana Juvvadi <juvvadi@lekha.org>
To: edw@detel.com
Cc: Ramana Juvvadi <juvvadi@lekha.org>,
        "Mr. Arlington Hewes" <tpcadmin@info.tpc.int>,
        Dave Wreski <dave@nic.com>, Steve Williams <steve@genie96.com>,
        flexfax@sgi.com
Subject: Re: flexfax: Hylafax rpm issues 
Sender: owner-flexfax@celestial.com

Ed Weinberg writes:


 > Here is the reasponse:
 > Warning, No "hylafax" service definition, using default 4559/tcp.
 > Can not reach server at host "localhost", port 4559.
 > (end of response)

Check that following entries are present


/etc/inetd.conf

	hylafax stream  tcp     nowait  fax     /usr/sbin/hfaxd hfaxd -I

/etc/services

	hylafax 4559/tcp # HylaFAX client-server protocol
	snpp    444/tcp         # Simple Network Paging Protocol

Note that rpm post-install script tries to add these entries. If
these entries are already present the script leaves them as
they are. If you have entries from previous installation,
make sure they match this. Or if you don't want to bother, simply do 
	rpm -e hyalfax
	rpm -i hyalafax

The rpm -e runs a pre-uninstall script which removes all these 
entries. rpm -i just adds them back.

Make sure that there is a symbolic link /dev/modem
to one of /dev/cua*

Now look at the file /var/spool/hyla/etc/config.modem. This is 
It is more than likely that it doesn't match all the parameters of your
modem

	(1) Do a 'cu -l modem' and find out as much info about
		your modem as possible with the commands

			ATI0 to ATI6
			AT+FCLASS?
			AT+FMDL?
			AT+FMFR?
			AT+FCR?

	(2) Now look at /var/spool/hyla/config. Modem parameters for
		some of the chipset are listed there. Pick what you
		think are the right set of parameters.

	(3) Now edit /var/spool/hyla/config.modem.  For a full listing
		of config parameters, do a 'man config'. Not all the 
		parameters need to be listed in the config file.
		Most of the defaults work. Pay particular
		attention to

			ModemType
			ModemRate
			ModemFlowControl
			ModemSendFillOrder
			ModemRecvFillOrder
			ModemWaitForConnect
			Class2RecvDataTrigger

	     Some trial and error on these parameters should work. 


			
This is not foolproof method. But I hope it takes you farther than
before.

Ramana




	 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Steve Williams <steve@genie96.com>
Subject: Re: flexfax: Hylafax rpm issues
To: juvvadi@lekha.org (Ramana Juvvadi)
Date: Wed, 22 Oct 1997 14:48:37 -0600 (MDT)
Cc: edw@detel.com, juvvadi@lekha.org, tpcadmin@info.tpc.int, dave@nic.com,
        steve@genie96.com, flexfax@sgi.com
Read-Receipt-To: steve@genie96.com
Sender: owner-flexfax@celestial.com

> 
> Ed Weinberg writes:
> 
> 
>  > Here is the reasponse:
>  > Warning, No "hylafax" service definition, using default 4559/tcp.
>  > Can not reach server at host "localhost", port 4559.
>  > (end of response)
> 
> Check that following entries are present
> 
> 
> /etc/inetd.conf
> 
> 	hylafax stream  tcp     nowait  fax     /usr/sbin/hfaxd hfaxd -I
> 

The preferred way of starting hfaxd is by a /etc/rc.local script or some
such thing.  It does work running from inetd, but over the years, it has
faded out in use.

Probably because of the necessity for inetd to fork off a hfaxd process for 
every fax.  On high volume machines, this is not good.  For home use, 
inetd is probably fine.


> /etc/services
> 
> 	hylafax 4559/tcp # HylaFAX client-server protocol
> 	snpp    444/tcp         # Simple Network Paging Protocol
> 

You should also include 4557 as "fax" or some such thing in /etc/services
as that is the "old" HylaFAX port.  There is still some "client" software
that will be looking for this port.

> Note that rpm post-install script tries to add these entries. If
> these entries are already present the script leaves them as
> they are. If you have entries from previous installation,
> make sure they match this. Or if you don't want to bother, simply do 
> 	rpm -e hyalfax
> 	rpm -i hyalafax
> 
> The rpm -e runs a pre-uninstall script which removes all these 
> entries. rpm -i just adds them back.
> 
> Make sure that there is a symbolic link /dev/modem
> to one of /dev/cua*
> 

Humm, is this a common practice in the Linux world?

MANY MANY MANY problems are caused by non-cooperative programs attempting
to use UUCP locking to prevent contention to the modem device.  

If you are advocating a link to /dev/modem, then it MUST be clear that 
PPP, minicom, cu, uucp, etc ALL USE /dev/modem as the device.  If only one
of those applications uses /dev/cua* as their device, the UUCP locking
will fail miserably.  This should be VERY CLEAR in your documentation.

Also, the LOCATION of the uucp lock files, and the TYPE of the UUCP 
lockfiles, though I would hope that RedHat has at least dictated those
two critical paramaters.

> Now look at the file /var/spool/hyla/etc/config.modem. This is 
> It is more than likely that it doesn't match all the parameters of your
> modem
> 

People should be running a faxaddmodem to add the modem to the system
( as per previous discussions ).  If it is hanging, one trick that I
have found to work ( *** modem dependent *** ) is to cu to the modem
( or minicom ), & set the modem to always assert CD.  This can usually
be accomplished by the AT&C0 command, followed by AT&W, though these
commands are modem specific.  Then, run faxaddmodem.  It may work this
time.
 
[snip]

-- 
	Steve Williams, Calgary, Alberta, Canada
	Genie Computer Systems Inc.
	steve@genie96.com

"A man doesn't begin to attain wisdom until he recognizes that he is 
 no longer indespensable."
- Admiral Richard E. Byrd ( 1888-1957 )

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 17:37:07 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Ramana Juvvadi <juvvadi@lekha.org>
Cc: Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: Hylafax rpm issues 
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Wed, 22 Oct 1997, Ramana Juvvadi wrote:

> Ed Weinberg writes:
> 
> 
>  > Here is the reasponse:
>  > Warning, No "hylafax" service definition, using default 4559/tcp.
>  > Can not reach server at host "localhost", port 4559.
>  > (end of response)
> 
> Check that following entries are present
> 
> 
> /etc/inetd.conf
> 
> 	hylafax stream  tcp     nowait  fax     /usr/sbin/hfaxd hfaxd -I

Bad ferret! No biscuit! The HylaFAX distribution should use the
"hylafax" script in the SysV init scripts to initiate the server, by
putting it into (as I remember) /etc/rc.2d and symbolic links to the
appropriate subdirectories there as Shylafax and Khylafax.

> /etc/services
> 
> 	hylafax 4559/tcp # HylaFAX client-server protocol
> 	snpp    444/tcp         # Simple Network Paging Protocol

You left out:
	fax	4557/tcp # Old Flexfax client-server protocol

> The rpm -e runs a pre-uninstall script which removes all these 
> entries. rpm -i just adds them back.
> 
> Make sure that there is a symbolic link /dev/modem
> to one of /dev/cua*

Again: /dev/cua* is not correct. You should use /dev/ttyS*, since the
main Linux developers (such as Ted T'so at MIT) believe that the
split-device trick used on /dev/ttyS* and /dev/cua* is a piece of
unspeakable mucous secretion and planned (at last question) to
discontinue its inclusion.

> This is not foolproof method. But I hope it takes you farther than
> before.

The effort is cool. Do you think you could accomodate my suggestions?

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNE5x9z/+ItycgIJRAQGpJgQAk2bzSU/cHGRap6NuuHG5weQZ6Z4NuV2e
txVfYLXbGbid79AtD1/QKfXJr2x5qIvc83VMFrPFndf5mU1fZWg2pcwDOcJv1MS5
jtck2/bAJvsfW9tcRhD4JI+ID39DBW//Xahqqq7JPtCJ7QbXjQdqWX7qUShvmUA4
KEzmvewt/9A=
=WR33
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: edw@detel.com (Ed Weinberg)
To: Ramana Juvvadi <juvvadi@lekha.org>
Cc: Ramana Juvvadi <juvvadi@lekha.org>,
        "Mr. Arlington Hewes" <tpcadmin@info.tpc.int>,
        Dave Wreski <dave@nic.com>, Steve Williams <steve@genie96.com>,
        flexfax@sgi.com
Subject: Re: flexfax: Hylafax rpm issues 
Date: Wed, 22 Oct 1997 21:58:35 GMT
Reply-To: edw@detel.com
Sender: owner-flexfax@celestial.com

On Wed, 22 Oct 1997 16:15:10 -0400, Ramana Juvvadi <juvvadi@lekha.org>
wrote:

>Ed Weinberg writes:
>
>
> > Here is the reasponse:
> > Warning, No "hylafax" service definition, using default 4559/tcp.
> > Can not reach server at host "localhost", port 4559.
> > (end of response)
>
>Check that following entries are present
Got everything in.  Now when I issue the command:
#sendfax -d 912033674005 /etc/hosts
I get:
Service not available, remote server closed connection

Am I issueing the correct command to send /etc/hosts to 9+1+telephone
number?  Is there anyplace I need to tell Hylafax that it needs to
send a 9 first?

  --  Ed Weinberg,
      Detel, Inc., An Internet Presence Provider
      edw@detel.com
      www.detel.com/
      www.serverking.com
      www.q5.com/  <-- find someone to CoolTalk or chat with here

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 18:05:10 -0400
From: Ramana Juvvadi <juvvadi@lekha.org>
To: Steve Williams <steve@genie96.com>
Cc: juvvadi@lekha.org (Ramana Juvvadi), edw@detel.com, tpcadmin@info.tpc.int,
        dave@nic.com, flexfax@sgi.com
Subject: Re: flexfax: Hylafax rpm issues
Sender: owner-flexfax@celestial.com

Steve Williams writes:

 > MANY MANY MANY problems are caused by non-cooperative programs attempting
 > to use UUCP locking to prevent contention to the modem device.  
 > 
 > If you are advocating a link to /dev/modem, then it MUST be clear that 
 > PPP, minicom, cu, uucp, etc ALL USE /dev/modem as the device.  If only one
 > of those applications uses /dev/cua* as their device, the UUCP locking
 > will fail miserably.  This should be VERY CLEAR in your documentation.

Redhat installation makes /dev/modem symbolic link at the time of 
installation. And yes, all the modem related programs like minicom do come
with /dev/modem as the default. 

I know that there is a contoversy on /dev/modem vs /dev/cua* 
( or /dev/ttS*). People have their preferences and I don't want 
to get embroiled in that. Whichever I choose I displease the
other camp. I just chose /dev/modem because that is what
is the default on Redhat ( I am willing to be corrected on this).

Ramana

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 18:15:45 -0400
From: Ramana Juvvadi <juvvadi@lekha.org>
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
Cc: Ramana Juvvadi <juvvadi@lekha.org>, Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: Hylafax rpm issues 
Sender: owner-flexfax@celestial.com

Nico Garcia writes:

 > Bad ferret! No biscuit! The HylaFAX distribution should use the
 > "hylafax" script in the SysV init scripts to initiate the server, by
 > putting it into (as I remember) /etc/rc.2d and symbolic links to the
 > appropriate subdirectories there as Shylafax and Khylafax.

The rpm already does this. However, the script starts only faxq
process. I put the hfaxd in the inetd.conf. I can easily start both
the the daemons from the same file. As I see inetd.conf soln
is for lowq volume fax servers. Whereas a standalone daemon is
for high volume fax servers. 


 > Again: /dev/cua* is not correct. You should use /dev/ttyS*, since the
 > main Linux developers (such as Ted T'so at MIT) believe that the
 > split-device trick used on /dev/ttyS* and /dev/cua* is a piece of
 > unspeakable mucous secretion and planned (at last question) to
 > discontinue its inclusion.

I will take your word for it. However, I am not deciding one way or
other in the rpm script. I just check if /dev/modem exists and 
and advise the user to have /dev/modem.


 > The effort is cool. Do you think you could accomodate my suggestions?

Sure. That is easy. 

Ramana

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 00:43:03 -0400
From: Ramana Juvvadi <juvvadi@lekha.org>
To: Matthew.Rice@ftlsol.com
Cc: flexfax@sgi.com
Subject: Re: flexfax: Hylafax rpm issues
Sender: owner-flexfax@celestial.com

Matthew.Rice@ftlsol.com writes:
 > > Check that following entries are present
 > > 
 > > /etc/inetd.conf
 > > 	hylafax stream  tcp     nowait  fax     /usr/sbin/hfaxd hfaxd -I
 > 
 > Is the RPM adding this line to inetd?  I think that most users would
 > rather see hfaxd run as a daemon.  Some simple rc scripts would be great.

Since this is suggested by more than one person, I'll move it to 
a standalone demon in the next release.


 > Also, is the inetd entry the reason that you recommended users reboot
 > after the install?  If you keep the entry then you can make inetd
 > reread it's configuration file with:
 >         kill -HUP `cat /var/run/inetd.pid`
 > 

The reason I recommended a reboot was for the faxgetty entry
in /etc/inittab. Is there anyway to make the init process read 
/etc/inittab?

Ramana

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 01:25:18 -0400 (EDT)
From: Dave Wreski <dave@nic.com>
To: "Mr. Arlington Hewes" <tpcadmin@info.tpc.int>
Subject: Re: flexfax: Hylafax rpm issues
Cc: (Ramana Juvvadi) <juvvadi@lekha.org>, flexfax@sgi.com,
        Steve Williams <steve@genie96.com>
Sender: owner-flexfax@celestial.com


On 22-Oct-97 "Mr. Arlington Hewes" wrote:
>>>>>> On Wed, 22 Oct 1997, "DW" == Dave Wreski wrote:
> 
>   +> My second question would be..., are you willing to support it yourself
>   +> for the next eternity?
> 
>   DW> Are you actually a linux user?  Everyone knows the package maintainer
>   DW> is simply the package maintainer, not the content.
> 
> I must agree with Steve here, Ramana has set the precedent of deviating from
> traditional HylaFAX build conventions, even to the point of renaming one
> of the scripts. In effect, he has become the content maintainer. There's
> no telling where this will lead. As a RedHat Linux user (before you start
> trying to drag me through the mud), I'm concerned that the 'official' RPM
> will be different enough from the rest of the packages that it will not be
> possible for us to support newbies who speak up in this forum.

Ok, first I didn't mean to be hasty in passing judgement.  Sorry if my tone was
harsh.  I was very focused on Steve's points about it seemingly not being worth
the effort for various reasons...

Anyway, I haven't looked at the RPM itself; I already have HylaFAX installed. 
I really didn't realize he changed as much as he did in the RPM.  I certainly
would not suggest leaving programs out, for instance.  This certainly becomes
the package maintainers responsiblity at that point to be bound to maintaining
it.

As far as the /usr vs /usr/local, that is a Linux Filesystem Standard, not
specifically a RedHat standard.  The FSSTND states that after a new
installation, /usr/local should be emtpy, except for the directory names. 
Ramana also intended this to go into the stock RH configuration.

Incidentally, I really doubt that will ever happen, because of the nature and
difficulty of installing it currently.

> Changes such as renaming xferstats should be addressed in this forum, and 
> changed in the master distribution, if justified.

Again, I shouldn't have assumed they already were before he built the package,
but I do not discount his efforts at making it one step closer.  Its certainly
not time to be announced to general public.

> I support much of what Ramana has done. In particular, sticking to RPM 
> tradition and wishing to be included in the next RedHat release are 
> admirable ambitions, and will do much to further the cause of us HylaFAX 
> evangelists. However, I do feel he's being a bit reactionary in making 
> sweeping changes based on minimal discussion.

As far as the /usr/hyla, which is certainly silly.  /var/spool/fax should
remain.  Possibly ./fax/hylafax and ./fax/otherfax, etc.

I also think changing the names are not worthwhile.  Possibly
/usr/share/hylafax, which is also a FSSTND, should be adopted to Linux-specific
installations.  This directory contains arch-independant data, like ghostscript
fonts, system-wide config information, etc..

See ya on the RH list.
Dave

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 02:09:55 -0400
From: Ramana Juvvadi <juvvadi@lekha.org>
To: Dave Wreski <dave@nic.com>
Cc: "Mr. Arlington Hewes" <tpcadmin@info.tpc.int>,
        (Ramana Juvvadi) <juvvadi@lekha.org>, flexfax@sgi.com,
        Steve Williams <steve@genie96.com>
Subject: Re: flexfax: Hylafax rpm issues
Sender: owner-flexfax@celestial.com

Dave Wreski writes:

 > I really didn't realize he changed as much as he did in the RPM.  I certainly
 > would not suggest leaving programs out, for instance.  This certainly becomes
 > the package maintainers responsiblity at that point to be bound to maintaining
 > it.


For the record, I didn't leave out anything. All I said was it may be
easier to edit the config files by hand than going through
faxaddmodem. To that end I supplied default config files
so that users can get away just by changing as few parameters as possible.
The whole argument was about my installation documentation than
anything I put into the RPM.


 > As far as the /usr/hyla, which is certainly silly.  /var/spool/fax should
 > remain.  Possibly ./fax/hylafax and ./fax/otherfax, etc.

I didn't use  /usr/hyla. I suggested /var/spool/hyla and
/usr/lib/hyla. Basically I want to replace 'fax' with
something more specific like hyla ( or hylafax).  

 > 
 > I also think changing the names are not worthwhile.  Possibly
 > /usr/share/hylafax, which is also a FSSTND, should be adopted to Linux-specific

Good point. I should have used /usr/share instead of /usr/lib as there
is no binary data there.

Ramana

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthew.Rice@ftlsol.com
X-Authentication-Warning: hudson.ftlsol.com: matt set sender to Matthew.Rice using -f
Subject: Re: flexfax: Hylafax rpm issues
To: juvvadi@lekha.org (Ramana Juvvadi)
Date: Thu, 23 Oct 1997 08:14:06 -0400 (EDT)
Cc: Matthew.Rice@ftlsol.com, flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> The reason I recommended a reboot was for the faxgetty entry
> in /etc/inittab. Is there anyway to make the init process read 
> /etc/inittab?

It does mention the telinit command in the init(8) manpage which
claims to do the trick.

I guess `kill -HUP 1` isn't politically correct any longer.  Although,
until I read the manpage this morning, this is the way that I've been
using...

Regards,
-- 
Matthew Rice                             e-mail: matthew.rice@ftlsol.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 09:47:01 -0400 (EDT)
From: Rob Newberry <rob@eats.com>
To: Ramana Juvvadi <juvvadi@lekha.org>
cc: flexfax@sgi.com
Subject: Re: flexfax: Hylafax rpm issues
Sender: owner-flexfax@celestial.com

On Thu, 23 Oct 1997, Ramana Juvvadi wrote:

> The reason I recommended a reboot was for the faxgetty entry
> in /etc/inittab. Is there anyway to make the init process read 
> /etc/inittab?

It works the same way -- send it a HUP signal.  Though there's no
"init.pid" file, its pid is always 1.  So:

	kill -HUP 1

results in (from /var/log/messages):

	Oct 23 11:01:45 molly init: Re-reading inittab

That's what you wanted, right?

Rob

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 11:26:56 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Ed Weinberg <edw@detel.com>
Cc: Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: Hylafax rpm issues 
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

> 
> Currently I am try ing to use sendfax.  The telephone line attached to
> the modem requires a "9" to dial out and to dial my fax number for
> testing I need to dial a "1".   here is what I tried:
> #sendfax -d 912033674005 /etc/hosts

You probably want to modify your dialrules file, by changing the end of it
to look like this. This will add a "9" in front of full length phone numbers,
without doing it to local shorter extension numbers.

  ^[+]			= ${IDPrefix}		! international call
  ^[0-9]{7,}$		= 9,&		! add 9 for outside calls
  ]

> Here is the reasponse:
> Warning, No "hylafax" service definition, using default 4559/tcp.
> Can not reach server at host "localhost", port 4559.
> (end of response)

hfaxd is not running. Run "faxstat" and see what you get, and check
for hfaxd entried in your /etc/inetd.conf.

> Can you suggest specific docs to read?

http://www.vix.com/hylafax/, and the Linux or x86 notes at
ftp://ftp.sgi.com/sgi/fax/binary/.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNE9ssj/+ItycgIJRAQFEnAP9E+Rl+IZ/8rN4b8ffmBqYH5WX+mW1G7v/
lVJaC51vP1JJgskQxmu2ib9BJ2eW4rpqoSjRKFGSc6tteevXVasdBIOKKDtJKlWq
YguAlauyFx/O6y7g2AjKpAq+dqZAgtvm8HEcbpMLdT7EJvKps/vbKGvrvNg75iXn
i+ygqGoJCTc=
=mA39
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Steve Williams <steve@genie96.com>
Subject: Re: flexfax: Hylafax rpm issues
To: videogram@earthlink.net
Date: Thu, 23 Oct 1997 09:29:54 -0600 (MDT)
Cc: flexfax@sgi.com
Read-Receipt-To: steve@genie96.com
Sender: owner-flexfax@celestial.com

> 
> Ramana Juvvadi wrote:
> > 
> > I hope with this rpm release, hylafax is easier than before to setup.
> Thanks For the rpm we were looking at Hylafax as an e-mail to fax
> solution, but were afraid it would be support hog due to the large
> number of questions regarding installation and configuration.Your work
> should make life easier.

If you are looking for a robust, professional solution, HylaFAX is the 
software, but it is only as good as the modem that you plug into it.

It will still be a support hog if you cheap out on the modems.   As it
stands now, there are probably only 3 modems worth considering, MultiTech
USR Courier ( ***NOT*** Sportster ), and maybe Zyxel's.  There maybe more,
but those are the 3 that come to mind as people having the least problems
with.

Anything else, and yes, you are looking at no end of headache's.

Good Luck.
-- 
	Steve Williams, Calgary, Alberta, Canada
	Genie Computer Systems Inc.
	steve@genie96.com

"A man doesn't begin to attain wisdom until he recognizes that he is 
 no longer indespensable."
- Admiral Richard E. Byrd ( 1888-1957 )

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 11:35:33 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Ramana Juvvadi <juvvadi@lekha.org>
Cc: Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: Hylafax rpm issues
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Thu, 23 Oct 1997, Ramana Juvvadi wrote:

>  > Also, is the inetd entry the reason that you recommended users reboot
>  > after the install?  If you keep the entry then you can make inetd
>  > reread it's configuration file with:
>  >         kill -HUP `cat /var/run/inetd.pid`
>  > 
> 
> The reason I recommended a reboot was for the faxgetty entry
> in /etc/inittab. Is there anyway to make the init process read 
> /etc/inittab?

Under SunOS, it is "kill -HUP 1", done as root. You can also examine
the "hylafax" script to examine the syntax of the "killall" command,
which might also do the job (can't swear to it, don't have a Linux box
in hand to check). Sending a HUP to the init process causes it to
re-read its configuration files, similarly to httpd and syslog. This
is how you modify serial line getty's in general.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNE9uvT/+ItycgIJRAQHocQP8DVrwgNxelohb2ShF/TCfFa1u9UfOH/5i
wPUsa9qEe2MdKWit3GxQgm26u0YQwhNLDAGMjhYCCk6dgfskRcJP6cLtBO8iwvGV
zcwr73v28aSqr1IVac2MujsbzVoRO72Q8ibt9JVViNg/YRkq8iqZZ0CYMCfe4cDg
cklaS5GwtFU=
=hQBs
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 13:46:31 -0400 (EDT)
From: Marty Kirkman <jkirkman@bellsouth.net>
Subject: Re: flexfax: Hylafax rpm issues
To: Rob Newberry <rob@eats.com>
Cc: Ramana Juvvadi <juvvadi@lekha.org>, flexfax@sgi.com
Sender: owner-flexfax@celestial.com

init q

 makes SCO Openserver reread inittab. Do

    man init

 on your system and it should tell you if it will work for you! 

Marty

On Thu, 23 Oct 1997, Rob Newberry wrote:

> On Thu, 23 Oct 1997, Ramana Juvvadi wrote:
> 
> > The reason I recommended a reboot was for the faxgetty entry
> > in /etc/inittab. Is there anyway to make the init process read 
> > /etc/inittab?
> 
> It works the same way -- send it a HUP signal.  Though there's no
> "init.pid" file, its pid is always 1.  So:
> 
> 	kill -HUP 1
> 
> results in (from /var/log/messages):
> 
> 	Oct 23 11:01:45 molly init: Re-reading inittab
> 
> That's what you wanted, right?
> 
> Rob
> 
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 14:58:26 -0500 (CDT)
From: Bill Suetholz <bill@centonline.com>
To: Ramana Juvvadi <juvvadi@lekha.org>
cc: "Mr. Arlington Hewes" <tpcadmin@info.tpc.int>, Dave Wreski <dave@nic.com>,
        Steve Williams <steve@genie96.com>, flexfax@sgi.com
Subject: Re: flexfax: Hylafax rpm issues 
Sender: owner-flexfax@celestial.com

Is there any way in the install phase of a RPM to ask for destination
locations for the runtime programs?  If so maybe something could be done
about the changing from "fax" to "hyla".  Some(Most?) installers would
like the option of being able to control the destination of add on
packages like HylaFAX.

Bill Suetholz
 
>On Wed, 22 Oct 1997, Ramana Juvvadi wrote:
>
> Mr. Arlington Hewes writes:
> 
> 
>  > tradition and wishing to be included in the next RedHat release are 
>  > admirable ambitions, and will do much to further the cause of us HylaFAX 
>  > evangelists. However, I do feel he's being a bit reactionary in making 
>  > sweeping changes based on minimal discussion.
> 
> The only "content" change I made is xferstat --> xferfaxstat
> xferlog --> xferfaxlog because it is clashing with wu-ftpd. 
> If it is addressed in the next version or if somebody
> has a better suggestion I am willing to change it.  
> 
> Ramana
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 15:03:01 -0500 (CDT)
From: Bill Suetholz <bill@centonline.com>
To: Ramana Juvvadi <juvvadi@lekha.org>
cc: flexfax@sgi.com, Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Hylafax rpm issues 
Sender: owner-flexfax@celestial.com

Suggest you exchange email with Matthias.Apitz@SOFTCON.de about the
xferstat name change to see about getting it changed in the next release. 
That is what most of the people are complaining about.  The arbitrary
picking of a new name for just one release of HylaFAX, without any real
prior discussion about the issue.  Same thing goes for the fax to hyla
directory change.

Bill Suetholz 

>On Wed, 22 Oct 1997, Ramana Juvvadi wrote: 
>
> Mr. Arlington Hewes writes:
> 
> 
>  > tradition and wishing to be included in the next RedHat release are 
>  > admirable ambitions, and will do much to further the cause of us HylaFAX 
>  > evangelists. However, I do feel he's being a bit reactionary in making 
>  > sweeping changes based on minimal discussion.
> 
> The only "content" change I made is xferstat --> xferfaxstat
> xferlog --> xferfaxlog because it is clashing with wu-ftpd. 
> If it is addressed in the next version or if somebody
> has a better suggestion I am willing to change it.  
> 
> Ramana
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 16:22:05 -0400
From: Ramana Juvvadi <juvvadi@lekha.org>
To: Bill Suetholz <bill@centonline.com>
Cc: Ramana Juvvadi <juvvadi@lekha.org>,
        "Mr. Arlington Hewes" <tpcadmin@info.tpc.int>,
        Dave Wreski <dave@nic.com>, Steve Williams <steve@genie96.com>,
        flexfax@sgi.com
Subject: Re: flexfax: Hylafax rpm issues 
Sender: owner-flexfax@celestial.com

Bill Suetholz writes:
 > Is there any way in the install phase of a RPM to ask for destination
 > locations for the runtime programs?  If so maybe something could be done
 > about the changing from "fax" to "hyla".  Some(Most?) installers would
 > like the option of being able to control the destination of add on
 > packages like HylaFAX.
 > 

That is called a relocatable rpm. While it is possible, the Makefile
and all the shell scripts should be written with rpm relocation in 
mind. Hylafax gives options at the compile time where the directories
should go. Making it relocatable involves changing the Makefile and
perhaps some scripts too. I want to limit my role to compiling
Hylafax with configure --with-<whatever-option>. 

To summarize, it can probably be done but some effort is required.

Ramana

 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: Hylafax rpm issues
To: raoul@cirl.meei.harvard.edu (Nico Garcia)
Date: Sat, 25 Oct 1997 13:44:36 +0100 (BST)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> 
> Freeware that demands top of the line modems is not freeware.

This smacks of "looking the gift horse in the mouth".

The term "free lunch" also comes to mind in this debate.  To some extent,
people like Red Hat are making money out of other people's free software.
I'm not sure what payment in kind they make to compensate, but it
is very difficult for commercial organisations to justify paying for
free material whether in kind or otherwise, when the accountants and
shareholders start to look.

One of the things that commercial software providers sell is support (even
though many seem to try and delegate it to their distributors).  Many use
freeware (e.g. SCO use sendmail, and other Berekely software) as the basis
of their products.  In fact, support is one of the major costs to commercial
software sellers.

Traditional freeware users, on Unix, have been people who could support
themselves, and are as likely to contribute patches to fix problems as to
ask for help in fixing them.

On the other hand, a lot of Linux users are from a plug and play mentality,
and Red Hat are exploiting this, by making it easier and easier to install
Linux without having enough knowledge to maintain it.  As a result, Red Hat,
etc. get the benefits, but people on the traditional support groups start
getting large numbers of really basic questions that someone who is using
the software ought to be able to answer themselves before they consider 
usinng anything except a commercially supported package.

Although I use Linux myself, I see a significant trend towards very low
quality questions being asked by Linux users, which, in my view are the
moral responsibility of the Linux packager, not of the original freeware
author.

Generally, the price of freeware is in the cost of supporting it yourself
and the price of cheap modems, as well as reliability issues, is in having
to spend time working round problems or money paying others to do so.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Sat, 25 Oct 1997 14:10:10 -0400
From: Ramana Juvvadi <juvvadi@lekha.org>
To: David Woolley <david@djwhome.demon.co.uk>
Cc: raoul@cirl.meei.harvard.edu (Nico Garcia), flexfax@sgi.com
Subject: Re: flexfax: Hylafax rpm issues
Sender: owner-flexfax@celestial.com

David Woolley writes:


 > On the other hand, a lot of Linux users are from a plug and play mentality,
 > and Red Hat are exploiting this, by making it easier and easier to install
 > Linux without having enough knowledge to maintain it.  As a result, Red Hat,
 > etc. get the benefits, but people on the traditional support groups start

You are essentially saying that let us keep installation of hylafax
( or all freeware) purposely difficult, so that it doesn't 
attract every Tom, Dick, and Harry, who will bug the original
authors with stupid questions. 

I don't see why the original author always has to answer the 
question. They can simply answer the more interesting questions
leaving all the basic questions to  other users. Then there are  
frequently asked questions. After all, I don't see Linus Torvalds
answering every question on Linux on the usenet groups.
I don't see freeware users expecting lot of handholding
from the original authors. A lot of support comes
from fellow users.

To take the particular case of hylafax itself, 'faxsetup' 
the way it is because authors perceive it as easy enough,
and because they think it is the right thing to do.
I don't think it is designed to weed out dumb users.

Ramana

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: Hylafax rpm issues
To: juvvadi@lekha.org (Ramana Juvvadi)
Date: Sun, 26 Oct 1997 15:48:01 +0000 (GMT)
Cc: raoul@cirl.meei.harvard.edu, flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> 
> 
>  > On the other hand, a lot of Linux users are from a plug and play mentality,
>  > and Red Hat are exploiting this, by making it easier and easier to install
>  > Linux without having enough knowledge to maintain it.  As a result, Red Hat,
>  > etc. get the benefits, but people on the traditional support groups start
> 
> You are essentially saying that let us keep installation of hylafax
> ( or all freeware) purposely difficult, so that it doesn't 
> attract every Tom, Dick, and Harry, who will bug the original
> authors with stupid questions. 

No.  I am saying that it is dishonest to make it easy to install and then
not also support it.  If, after you port it, you fend off all the FAQs
on the mailing list (and to some extent you, personally, are trying to do
this) that's OK, but if you don't put resources (which would be money
in a commercial environment) into supporting it, you are pushing your
costs onto someone else.  Saying that the packager is not responsible is
equivalent to passing the buck onto the mailing list.

Incidentally, high volumes of FAQs can make the original supporters of
a list abandon it, because it becomes difficult to extract the worthwhile
from the trivial.  This can have a double impact, because it means that
no-one is vetting answers for misunderstandings.

Microsoft style newsgroups tend to be even worse in this respect, multiple
cross posting to irrelevant groups and please email me I don't read the
group, and not checking the recent history for similar questions, seem
to be the norm, but Linux users seem to be coming somewhere in the middle.

There used to be a guideline not to ask questions in Linux about ports
of general internet software, but I feel that it might be an idea to
revise that position.

Incidentally, Linux has been getting a bad reputation in support newsgroups
for several years, not just since Red Hat became prominent.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
To: Ramana Juvvadi <juvvadi@lekha.org>
cc: Steve Williams <steve@genie96.com>, edw@detel.com, flexfax@sgi.com
Subject: flexfax: Re: RPM: document conversion script was not found 
Reply-To: Sam Leffler <sam@engr.sgi.com>
Date: Thu, 23 Oct 1997 18:09:08 -0700
From: Sam Leffler <sam@hyla.chez.sgi.com>
Sender: owner-flexfax@celestial.com

    To:  Steve Williams <steve@genie96.com>
    Subject:  Re: RPM: document conversion script was not found
    Cc:  edw@detel.com, flexfax@sgi.com
    Date: Thu, 23 Oct 1997 19:05:12 EDT
    From:  Ramana Juvvadi <juvvadi@lekha.org>

    Steve Williams writes:
    
     > called ps2fax, that should fire up ghostscript.  Either it's missing,
     > or what it is doing is wrong.  Try to run this program by hand.  There 
is
     > a man page ( I believe ) for ps2fax ( NOTE I believe you have to be in 
the
     > $FAXSPOOL directory when you run ps2fax )
    
    Actually, this was an error in my rpm. I should have included
    a symbolic link ps2fax --> ps2fax.gs in /var/spool/hyla/bin.
    I will fix it in the next release of the rpm.

The right solution (as I told you in direct email) is to use the faxsetup
script to do work such as configuring the symbolic link to the appropriate
ps2fax script.  Not only does faxsetup install this link but it also verifies
that the PSRIP is installed on the target machine and that the RIP is
properly configured (e.g. that Ghostscript has the necessary driver).

FWIW my binary distribution for IRIX that is at least as complete as
anything you are going to do requires that faxsetup be run to complete
an installation.
    
    The ModemClass: 2 was also introduced by me in the rpm. It will be
    removed in the next release. As a clueless user I typed 2 when 
    faxsetup asked for ModemClass. Basically I copied the config
    file faxsetup generated for me. 
    
    I have no complaints about faxsetup. However, the information
    that  faxsetup probes and finds out doesn't vary across
    Readhat Linux systems. It is the information that faxaddmodem
    is trying to find out that is likely to vary from 
    machine to machine. Unfortuantely, that is the information
    that is most difficult to find out by probing. I don't
    mean to complain about the program as such. I just think
    that it is difficult for any program to assure that
    probing a modem works reliably. I don't see how
    a programmer can test his program on hundreds of 
    models. 
    
faxsetup and faxaddmodem are designed to do many things to insure a
system is properly setup for operation.  While you may think that
faxsetup always does the same thing across your systems I believe that
you are wrong (consider above how faxsetup verifies that gs is built
and installed such that it supports HylaFAX's needs).  Further by not
simply invoking faxsetup to do it's thing you are missing out any future
improvements that might be made in this script.  Likewise for faxaddmodem.
The issue of faxsetup and/or faxaddmodem not working right for your system
is orthogonal to whether or not you should use it in your binary distribution.
Users will want (and need) to be able to run these scripts at times after
they've installed the software from the RPM so if they don't work then
need to do is fix 'em, not avoid them.

        Sam

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 21:55:06 -0400
From: Ramana Juvvadi <juvvadi@lekha.org>
To: Sam Leffler <sam@engr.sgi.com>
Cc: Ramana Juvvadi <juvvadi@lekha.org>, Steve Williams <steve@genie96.com>,
        edw@detel.com, flexfax@sgi.com
Subject: flexfax: Re: RPM: document conversion script was not found 
Sender: owner-flexfax@celestial.com

Sam Leffler writes:


 > faxsetup and faxaddmodem are designed to do many things to insure a
 > system is properly setup for operation.  While you may think that
 > faxsetup always does the same thing across your systems I believe that
 > you are wrong (consider above how faxsetup verifies that gs is built
 > and installed such that it supports HylaFAX's needs).  Further by not
 > simply invoking faxsetup to do it's thing you are missing out any future
 > improvements that might be made in this script.  Likewise for faxaddmodem.
 > The issue of faxsetup and/or faxaddmodem not working right for your system
 > is orthogonal to whether or not you should use it in your binary distribution.
 > Users will want (and need) to be able to run these scripts at times after
 > they've installed the software from the RPM so if they don't work then
 > need to do is fix 'em, not avoid them.


For the record, I already stated that I am advocating the rpm users to
run faxsetup and faxaddmodem. The installation notes that I am going
to put on te webpage will ask people to try hand configuring only if
faxaddmodem hangs or fails. I hope nobody would have serious
objections for that. I just don't think it is easy to write
faxaddmodem that will work reliably for everybody.  Some people will
have no choice but to do it by hand. The default configuration
files are meant for only those.

Ramana
 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 21 Oct 1997 11:28:31 -0600
From: Larry Pyeatt <pyeatt@cs.colostate.edu>
Organization: Colorado State University
To: flexfax@sgi.com
Subject: flexfax: MultiTech MT2834ZDXV recommendation?
Sender: owner-flexfax@celestial.com

Hello,
  I have given up on my USR Sportster and am looking for a modem that
works.
Has anyone used the MT2834ZDXV with hylafax?  If so, does it work well?

Thanks.

--
Larry D. Pyeatt                     All standard disclaimers apply.
pyeatt@cs.colostate.edu             Void where prohibited.
http://www.cs.colostate.edu/~pyeatt

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 21 Oct 1997 14:36:59 -0500 (CDT)
From: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
To: Larry Pyeatt <pyeatt@cs.colostate.edu>
cc: flexfax@sgi.com
Subject: Re: flexfax: MultiTech MT2834ZDXV recommendation?
Sender: owner-flexfax@celestial.com

I heard very good things about it from Dave from Ameritech. But he is now
trying to use an AT&T modem. Dave could you clarify this point for all of
us?

Why not use MultiTech?

Another point. Does anybody know of any fax modem that can detect the
tri-tone operator intercept tones, to distinguish between a busy/no answer
line and a wrong/disconnected/trunk busy/etc. number. Without this
feature, there is no way calls can be handled correctly.

Julio

On Tue, 21 Oct 1997, Larry Pyeatt wrote:

> Hello,
>   I have given up on my USR Sportster and am looking for a modem that
> works.
> Has anyone used the MT2834ZDXV with hylafax?  If so, does it work well?
> 
> Thanks.
> 
> --
> Larry D. Pyeatt                     All standard disclaimers apply.
> pyeatt@cs.colostate.edu             Void where prohibited.
> http://www.cs.colostate.edu/~pyeatt
> 
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 08:10:55 -0400 (EDT)
Organization: The Creative Group
From: Patrick Gardella <patrick@cre8tivegroup.com>
To: Larry Pyeatt <pyeatt@cs.colostate.edu>
Subject: RE: flexfax: MultiTech MT2834ZDXV recommendation?
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

I am using 8 MT2834ZDX modems are they have not given me a problem yet!  What
is the "v" mean?  It this the 28.8 -33.6 update?

Been running 24x7 for 3 months at an average of 2000 faxes outgoing. (I don't
allow any incoming.)

Patrick

On 21-Oct-97 Larry Pyeatt wrote:
>Hello,
>  I have given up on my USR Sportster and am looking for a modem that
>works.
>Has anyone used the MT2834ZDXV with hylafax?  If so, does it work well?
>
>Thanks.
>
>--
>Larry D. Pyeatt                     All standard disclaimers apply.
>pyeatt@cs.colostate.edu             Void where prohibited.
>http://www.cs.colostate.edu/~pyeatt

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 07:30:38 -0500 (EST)
From: Dave Lewis <dlewis@appn.ci.in.ameritech.com>
To: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
Cc: flexfax@sgi.com, Larry Pyeatt <pyeatt@cs.colostate.edu>
Subject: Re: flexfax: MultiTech MT2834ZDXV recommendation?
Sender: owner-flexfax@celestial.com

Sorry about being vague.  In our modem pool, we have one MultiTech 
(1432BA), one AT&T (with which I'm having the phase B/C comm. errors), 
and 7 USRobitics 14400's.  Of the nine modems, the MultiTech seems to 
have the least number of failed transmissions;  I wish all our modems 
were of this type.  Thanks, Dave Lewis


On Tue, 21 Oct 1997, Prof. Julio C. Spinelli wrote:

> I heard very good things about it from Dave from Ameritech. But he is now
> trying to use an AT&T modem. Dave could you clarify this point for all of
> us?
> 
> Why not use MultiTech?
> 
> Another point. Does anybody know of any fax modem that can detect the
> tri-tone operator intercept tones, to distinguish between a busy/no answer
> line and a wrong/disconnected/trunk busy/etc. number. Without this
> feature, there is no way calls can be handled correctly.
> 
> Julio
> 
> On Tue, 21 Oct 1997, Larry Pyeatt wrote:
> 
> > Hello,
> >   I have given up on my USR Sportster and am looking for a modem that
> > works.
> > Has anyone used the MT2834ZDXV with hylafax?  If so, does it work well?
> > 
> > Thanks.
> > 
> > --
> > Larry D. Pyeatt                     All standard disclaimers apply.
> > pyeatt@cs.colostate.edu             Void where prohibited.
> > http://www.cs.colostate.edu/~pyeatt
> > 
> > 
> > 
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
X-Authentication-Warning: guinness.chelsea.net: davidb owned process doing -bs
Date: Wed, 22 Oct 1997 09:43:57 -0400 (EDT)
From: David Birnbaum <davidb@chelsea.net>
To: Dave Lewis <dlewis@appn.ci.in.ameritech.com>
cc: flexfax@sgi.com
Subject: Re: flexfax: MultiTech MT2834ZDXV recommendation?
Sender: owner-flexfax@celestial.com

On Wed, 22 Oct 1997, Dave Lewis wrote:

> Sorry about being vague.  In our modem pool, we have one MultiTech 
> (1432BA), one AT&T (with which I'm having the phase B/C comm. errors), 
> and 7 USRobitics 14400's.  Of the nine modems, the MultiTech seems to 
> have the least number of failed transmissions;  I wish all our modems 
> were of this type.  Thanks, Dave Lewis

I can't be quite as happy - I have a USR Courier, a MultiTech MT2834ZDX,
and a Telebit WorldBlazer.  Of the three, the USR seems to be the most
stable.  The MT seems to, every 50-100 faxes or so, "time out" and I get
an error message "Unable to configure modem for fax use".  The next fax to
that modem goes fine, so its clearly a very strange and transient problem,
but in the meantime, the fax that died needs to be resubmitted because
HylaFax kills the job rather than reroute it to another modem.  Anybody
interested in rexamining that behavior?

I'm getting ready to install a bank of 8 USR 33.6 modems, though -
leftover from a dial-in pool upgrade, I hope to get it functioning in the
next few weeks.

David.

> On Tue, 21 Oct 1997, Prof. Julio C. Spinelli wrote:
> 
> > I heard very good things about it from Dave from Ameritech. But he is now
> > trying to use an AT&T modem. Dave could you clarify this point for all of
> > us?
> > 
> > Why not use MultiTech?
> > 
> > Another point. Does anybody know of any fax modem that can detect the
> > tri-tone operator intercept tones, to distinguish between a busy/no answer
> > line and a wrong/disconnected/trunk busy/etc. number. Without this
> > feature, there is no way calls can be handled correctly.
> > 
> > Julio
> > 
> > On Tue, 21 Oct 1997, Larry Pyeatt wrote:
> > 
> > > Hello,
> > >   I have given up on my USR Sportster and am looking for a modem that
> > > works.
> > > Has anyone used the MT2834ZDXV with hylafax?  If so, does it work well?
> > > 
> > > Thanks.
> > > 
> > > --
> > > Larry D. Pyeatt                     All standard disclaimers apply.
> > > pyeatt@cs.colostate.edu             Void where prohibited.
> > > http://www.cs.colostate.edu/~pyeatt
> > > 
> > > 
> > > 
> > 
> > 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 11:32:34 -0500 (CDT)
From: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
To: Patrick Gardella <patrick@cre8tivegroup.com>
cc: Larry Pyeatt <pyeatt@cs.colostate.edu>, flexfax@sgi.com
Subject: RE: flexfax: MultiTech MT2834ZDXV recommendation?
Sender: owner-flexfax@celestial.com

Dear Patrick,

Could you please provide me with the config file that you are using. I
would like to try the Multitech modem. Do you really send 2000 faxes per
day average? Whow! Are they all to the USA or international?

Thanks very much for this help,

Modem vendors will learn to market products that really work thaks to the
exchange of info that you are doing. Thanks again,

Julio

On Wed, 22 Oct 1997, Patrick Gardella wrote:

> I am using 8 MT2834ZDX modems are they have not given me a problem yet!  What
> is the "v" mean?  It this the 28.8 -33.6 update?
> 
> Been running 24x7 for 3 months at an average of 2000 faxes outgoing. (I don't
> allow any incoming.)
> 
> Patrick
> 
> On 21-Oct-97 Larry Pyeatt wrote:
> >Hello,
> >  I have given up on my USR Sportster and am looking for a modem that
> >works.
> >Has anyone used the MT2834ZDXV with hylafax?  If so, does it work well?
> >
> >Thanks.
> >
> >--
> >Larry D. Pyeatt                     All standard disclaimers apply.
> >pyeatt@cs.colostate.edu             Void where prohibited.
> >http://www.cs.colostate.edu/~pyeatt
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 12:47:41 -0400 (EDT)
Organization: The Creative Group
From: Patrick Gardella <patrick@cre8tivegroup.com>
To: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
Subject: RE: flexfax: MultiTech MT2834ZDXV recommendation?
Cc: flexfax@sgi.com, Larry Pyeatt <pyeatt@cs.colostate.edu>
Sender: owner-flexfax@celestial.com

I do send 2000+ a day.  Our record day is 4041 in 24 hours on 7 modems.  All the
faxes are 1 to 2 pages, no cover sheets and all are US (local) numbers.

Here is my config file for one of the modems:  (The FAXnumber and LocalId are 
commented out :)


Patrick
_________________

CountryCode:            1
AreaCode:               703
FAXNumber:              +xxxxx
LongDistancePrefix:     1
InternationalPrefix:    011
DialStringRules:        "etc/dialrules"
ServerTracing:          1
SessionTracing:         11
RecvFileMode:           0600
LogFileMode:            0600
DeviceMode:             0660
RingsBeforeAnswer:      0 
SpeakerVolume:          off
GettyArgs:              "std.%s"
LocalIdentifier:        "xxxx"
TagLineFont:            etc/lutRS18.pcf
TagLineFormat:          "From %%l|%c|Page %%p of %%t"
MaxRecvPages:           25
# 
# 
# Modem-related stuff: should reflect modem command interface
# and hardware connection/cabling (e.g. flow control).
#
ModemType:              Class2          # use class 2 interface
ModemRate:              38400           # lock rate for DCE-DTE communication
ModemFlowControl:       rtscts          # default
#
ModemHardFlowCmd:       AT&E4           # hardware flow control
ModemSoftFlowCmd:       AT&E5           # software flow control
ModemSetupDTRCmd:       AT&D3           # setup so DTR drop resets modem
ModemSetupDCDCmd:       AT&C1           # setup so DCD reflects carrier (or not)
ModemSetupAACmd:        AT+FAA=1        # enable adaptive-answer in class 2
#
# NB: some models get confused by the @
#
ModemDialCmd:           ATDT%s@         # T for tone dialing, @ for silence
#
Class2RecvDataTrigger:  "\022"          # character sent to modem to start recv


On 22-Oct-97 Prof. Julio C. Spinelli wrote:
>Dear Patrick,
>
>Could you please provide me with the config file that you are using. I
>would like to try the Multitech modem. Do you really send 2000 faxes per
>day average? Whow! Are they all to the USA or international?
>
>Thanks very much for this help,
>
>Modem vendors will learn to market products that really work thaks to the
>exchange of info that you are doing. Thanks again,
>
>Julio
>
>On Wed, 22 Oct 1997, Patrick Gardella wrote:
>
>> I am using 8 MT2834ZDX modems are they have not given me a problem yet!  What
>> is the "v" mean?  It this the 28.8 -33.6 update?
>> 
>> Been running 24x7 for 3 months at an average of 2000 faxes outgoing. (I don't
>> allow any incoming.)
>> 
>> Patrick
>> 
>> On 21-Oct-97 Larry Pyeatt wrote:
>> >Hello,
>> >  I have given up on my USR Sportster and am looking for a modem that
>> >works.
>> >Has anyone used the MT2834ZDXV with hylafax?  If so, does it work well?
>> >
>> >Thanks.
>> >
>> >--
>> >Larry D. Pyeatt                     All standard disclaimers apply.
>> >pyeatt@cs.colostate.edu             Void where prohibited.
>> >http://www.cs.colostate.edu/~pyeatt
>> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 12:05:44 -0500 (CDT)
From: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
To: David Birnbaum <davidb@chelsea.net>
cc: Dave Lewis <dlewis@appn.ci.in.ameritech.com>, flexfax@sgi.com
Subject: Re: flexfax: MultiTech MT2834ZDXV recommendation?
Sender: owner-flexfax@celestial.com

We received an email today telling us that somebody is sending 2000 faxes
a day with no problem with the MultiTech 2834ZDXV. His name is Patrick
Gardella. 

I propose that to sort this out both of you send to the list your
respective config files to see whether the difference is there. If not
please check on the exact firmware release dates by looking at the hylafax
logs.

I am using USR 33.6 now and I am having a failure rate of around 10%.

Thanks,

Julio

On Wed, 22 Oct 1997, David Birnbaum wrote:

> On Wed, 22 Oct 1997, Dave Lewis wrote:
> 
> > Sorry about being vague.  In our modem pool, we have one MultiTech 
> > (1432BA), one AT&T (with which I'm having the phase B/C comm. errors), 
> > and 7 USRobitics 14400's.  Of the nine modems, the MultiTech seems to 
> > have the least number of failed transmissions;  I wish all our modems 
> > were of this type.  Thanks, Dave Lewis
> 
> I can't be quite as happy - I have a USR Courier, a MultiTech MT2834ZDX,
> and a Telebit WorldBlazer.  Of the three, the USR seems to be the most
> stable.  The MT seems to, every 50-100 faxes or so, "time out" and I get
> an error message "Unable to configure modem for fax use".  The next fax to
> that modem goes fine, so its clearly a very strange and transient problem,
> but in the meantime, the fax that died needs to be resubmitted because
> HylaFax kills the job rather than reroute it to another modem.  Anybody
> interested in rexamining that behavior?
> 
> I'm getting ready to install a bank of 8 USR 33.6 modems, though -
> leftover from a dial-in pool upgrade, I hope to get it functioning in the
> next few weeks.
> 
> David.
> 
> > On Tue, 21 Oct 1997, Prof. Julio C. Spinelli wrote:
> > 
> > > I heard very good things about it from Dave from Ameritech. But he is now
> > > trying to use an AT&T modem. Dave could you clarify this point for all of
> > > us?
> > > 
> > > Why not use MultiTech?
> > > 
> > > Another point. Does anybody know of any fax modem that can detect the
> > > tri-tone operator intercept tones, to distinguish between a busy/no answer
> > > line and a wrong/disconnected/trunk busy/etc. number. Without this
> > > feature, there is no way calls can be handled correctly.
> > > 
> > > Julio
> > > 
> > > On Tue, 21 Oct 1997, Larry Pyeatt wrote:
> > > 
> > > > Hello,
> > > >   I have given up on my USR Sportster and am looking for a modem that
> > > > works.
> > > > Has anyone used the MT2834ZDXV with hylafax?  If so, does it work well?
> > > > 
> > > > Thanks.
> > > > 
> > > > --
> > > > Larry D. Pyeatt                     All standard disclaimers apply.
> > > > pyeatt@cs.colostate.edu             Void where prohibited.
> > > > http://www.cs.colostate.edu/~pyeatt
> > > > 
> > > > 
> > > > 
> > > 
> > > 
> > 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
X-Authentication-Warning: guinness.chelsea.net: davidb owned process doing -bs
Date: Wed, 22 Oct 1997 13:51:44 -0400 (EDT)
From: David Birnbaum <davidb@chelsea.net>
To: Patrick Gardella <patrick@cre8tivegroup.com>
cc: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>, flexfax@sgi.com,
        Larry Pyeatt <pyeatt@cs.colostate.edu>
Subject: RE: flexfax: MultiTech MT2834ZDXV recommendation?
Sender: owner-flexfax@celestial.com

I just checked my config file, and it is identical.  Here is a sample
failure trace:

Oct 02 19:47:36.55: [ 8646]: SESSION BEGIN 00001980 12126331423
Oct 02 19:47:36.55: [ 8646]: SEND FAX: JOB 2665 DEST 6331423 COMMID 00001980
Oct 02 19:47:36.55: [ 8646]: DELAY 2600 ms
Oct 02 19:47:39.15: [ 8646]: <-- [15:ATE0V1Q0S0=0H0\r]
Oct 02 19:47:39.40: [ 8646]: --> [14:ATE0V1Q0S0=0H0]
Oct 02 19:47:39.51: [ 8646]: --> [2:OK]
Oct 02 19:47:39.51: [ 8646]: <-- [21:ATS8=2S7=60&E4&D3&C1\r]
Oct 02 19:47:39.83: [ 8646]: --> [2:OK]
Oct 02 19:47:39.83: [ 8646]: <-- [12:AT+FCLASS=2\r]
Oct 02 19:47:40.15: [ 8646]: --> [2:OK]
Oct 02 19:47:40.15: [ 8646]: <-- [10:AT+FTBC=0\r]
Oct 02 19:47:40.47: [ 8646]: --> [2:OK]
Oct 02 19:47:40.47: [ 8646]: <-- [10:AT+FBOR=0\r]
Oct 02 19:47:40.79: [ 8646]: --> [2:OK]
Oct 02 19:47:40.79: [ 8646]: <-- [13:AT+FPHCTO=30\r]
Oct 02 19:47:41.10: [ 8646]: --> [2:OK]
Oct 02 19:47:41.11: [ 8646]: <-- [24:AT+FDCC=1,5,2,2,0,0,0,0\r]
Oct 02 19:47:41.43: [ 8646]: --> [2:OK]
Oct 02 19:47:41.43: [ 8646]: <-- [5:ATM0\r]
Oct 02 19:47:41.75: [ 8646]: --> [2:OK]
Oct 02 19:47:41.82: [ 8646]: <-- [12:AT+FCLASS=2\r]
Oct 02 19:47:42.23: [ 8646]: --> [2:OK]
Oct 02 19:47:42.24: [ 8646]: <-- [10:AT+FTBC=0\r]
Oct 02 19:47:42.64: [ 8646]: --> [2:OK]
Oct 02 19:47:42.65: [ 8646]: <-- [10:AT+FBOR=0\r]
Oct 02 19:47:43.05: [ 8646]: --> [2:OK]
Oct 02 19:47:43.06: [ 8646]: <-- [13:AT+FPHCTO=30\r]
Oct 02 19:47:43.46: [ 8646]: --> [2:OK]
Oct 02 19:47:43.47: [ 8646]: <-- [24:AT+FDCC=1,5,2,2,0,0,0,0\r]
Oct 02 19:47:43.88: [ 8646]: --> [2:OK]
Oct 02 19:47:43.89: [ 8646]: <-- [27:AT+FLID="CHELSEA NETWORKS"\r]
Oct 02 19:48:13.90: [ 8646]: MODEM <Timeout>
Oct 02 19:48:13.90: [ 8646]: SEND FAILED: Unable to configure modem for fax use
Oct 02 19:48:13.91: [ 8646]: SESSION END

Note the timeout.  I know there are some issues with setting the ID to a
non-numeric parameter, but the fact that it works 97 times out of a
hundred is what is so strange about it. 

USR users - does anybody actually use the USR chassis for faxing?  I have
a MP 8 that I was going to use, but all of the notes from people about USR
problems has me a bit nervous.

Incidently, I sent out 600-700 faxes a night using only two modems right
now....

David.

On Wed, 22 Oct 1997, Patrick Gardella wrote:

> I do send 2000+ a day.  Our record day is 4041 in 24 hours on 7 modems.  All the
> faxes are 1 to 2 pages, no cover sheets and all are US (local) numbers.
> 
> Here is my config file for one of the modems:  (The FAXnumber and LocalId are 
> commented out :)
> 
> 
> Patrick
> _________________
> 
> CountryCode:            1
> AreaCode:               703
> FAXNumber:              +xxxxx
> LongDistancePrefix:     1
> InternationalPrefix:    011
> DialStringRules:        "etc/dialrules"
> ServerTracing:          1
> SessionTracing:         11
> RecvFileMode:           0600
> LogFileMode:            0600
> DeviceMode:             0660
> RingsBeforeAnswer:      0 
> SpeakerVolume:          off
> GettyArgs:              "std.%s"
> LocalIdentifier:        "xxxx"
> TagLineFont:            etc/lutRS18.pcf
> TagLineFormat:          "From %%l|%c|Page %%p of %%t"
> MaxRecvPages:           25
> # 
> # 
> # Modem-related stuff: should reflect modem command interface
> # and hardware connection/cabling (e.g. flow control).
> #
> ModemType:              Class2          # use class 2 interface
> ModemRate:              38400           # lock rate for DCE-DTE communication
> ModemFlowControl:       rtscts          # default
> #
> ModemHardFlowCmd:       AT&E4           # hardware flow control
> ModemSoftFlowCmd:       AT&E5           # software flow control
> ModemSetupDTRCmd:       AT&D3           # setup so DTR drop resets modem
> ModemSetupDCDCmd:       AT&C1           # setup so DCD reflects carrier (or not)
> ModemSetupAACmd:        AT+FAA=1        # enable adaptive-answer in class 2
> #
> # NB: some models get confused by the @
> #
> ModemDialCmd:           ATDT%s@         # T for tone dialing, @ for silence
> #
> Class2RecvDataTrigger:  "\022"          # character sent to modem to start recv
> 
> 
> On 22-Oct-97 Prof. Julio C. Spinelli wrote:
> >Dear Patrick,
> >
> >Could you please provide me with the config file that you are using. I
> >would like to try the Multitech modem. Do you really send 2000 faxes per
> >day average? Whow! Are they all to the USA or international?
> >
> >Thanks very much for this help,
> >
> >Modem vendors will learn to market products that really work thaks to the
> >exchange of info that you are doing. Thanks again,
> >
> >Julio
> >
> >On Wed, 22 Oct 1997, Patrick Gardella wrote:
> >
> >> I am using 8 MT2834ZDX modems are they have not given me a problem yet!  What
> >> is the "v" mean?  It this the 28.8 -33.6 update?
> >> 
> >> Been running 24x7 for 3 months at an average of 2000 faxes outgoing. (I don't
> >> allow any incoming.)
> >> 
> >> Patrick
> >> 
> >> On 21-Oct-97 Larry Pyeatt wrote:
> >> >Hello,
> >> >  I have given up on my USR Sportster and am looking for a modem that
> >> >works.
> >> >Has anyone used the MT2834ZDXV with hylafax?  If so, does it work well?
> >> >
> >> >Thanks.
> >> >
> >> >--
> >> >Larry D. Pyeatt                     All standard disclaimers apply.
> >> >pyeatt@cs.colostate.edu             Void where prohibited.
> >> >http://www.cs.colostate.edu/~pyeatt
> >> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 14:21:09 -0400 (EDT)
Organization: The Creative Group
From: Patrick Gardella <patrick@cre8tivegroup.com>
To: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
Subject: Re: flexfax: MultiTech MT2834ZDXV recommendation?
Cc: flexfax@sgi.com, Dave Lewis <dlewis@appn.ci.in.ameritech.com>,
        David Birnbaum <davidb@chelsea.net>
Sender: owner-flexfax@celestial.com


On 22-Oct-97 Prof. Julio C. Spinelli wrote:
>We received an email today telling us that somebody is sending 2000 faxes
>a day with no problem with the MultiTech 2834ZDXV. His name is Patrick
>Gardella. 
>
>I propose that to sort this out both of you send to the list your
>respective config files to see whether the difference is there. If not
>please check on the exact firmware release dates by looking at the hylafax
>logs.

My config file was sent previously.  Here is a sample log file:

If this is not what you want, please let me know.

Note:  I am using a MultiTech 2834ZDX not a V model. 

Patrick  

Sep 30 00:42:15.33: [ 3273]: SESSION BEGIN 00033009 17032022250566
Sep 30 00:42:15.33: [ 3273]: SEND FAX: JOB 19647 DEST 202-225-0566 COMMID 000330
09
Sep 30 00:42:15.33: [ 3273]: DELAY 2600 ms
Sep 30 00:42:17.94: [ 3273]: <-- [15:ATE0V1Q0S0=0H0\r]
Sep 30 00:42:18.15: [ 3273]: --> [14:ATE0V1Q0S0=0H0]
Sep 30 00:42:18.15: [ 3273]: --> [2:OK]
Sep 30 00:42:18.15: [ 3273]: <-- [21:ATS8=2S7=60&E4&D3&C1\r]
Sep 30 00:42:18.32: [ 3273]: --> [2:OK]
Sep 30 00:42:18.32: [ 3273]: <-- [12:AT+FCLASS=2\r]
Sep 30 00:42:18.48: [ 3273]: --> [2:OK]
Sep 30 00:42:18.48: [ 3273]: <-- [10:AT+FTBC=0\r]
Sep 30 00:42:18.64: [ 3273]: --> [2:OK]
Sep 30 00:42:18.64: [ 3273]: <-- [10:AT+FBOR=0\r]
Sep 30 00:42:18.80: [ 3273]: --> [2:OK]
Sep 30 00:42:18.80: [ 3273]: <-- [13:AT+FPHCTO=30\r]
Sep 30 00:42:18.96: [ 3273]: --> [2:OK]
Sep 30 00:42:18.96: [ 3273]: <-- [24:AT+FDCC=1,5,2,2,2,0,0,0\r]
Sep 30 00:42:19.13: [ 3273]: --> [2:OK]
Sep 30 00:42:19.13: [ 3273]: <-- [5:ATM0\r]
Sep 30 00:42:19.29: [ 3273]: --> [2:OK]
Sep 30 00:42:19.29: [ 3273]: <-- [12:AT+FCLASS=2\r]
Sep 30 00:42:19.56: [ 3273]: --> [2:OK]
Sep 30 00:42:19.56: [ 3273]: <-- [10:AT+FTBC=0\r]
Sep 30 00:42:19.82: [ 3273]: --> [2:OK]
Sep 30 00:42:19.82: [ 3273]: <-- [10:AT+FBOR=0\r]
Sep 30 00:42:20.08: [ 3273]: --> [2:OK]
Sep 30 00:42:20.08: [ 3273]: <-- [13:AT+FPHCTO=30\r]
Sep 30 00:42:20.34: [ 3273]: --> [2:OK]
Sep 30 00:42:20.34: [ 3273]: <-- [24:AT+FDCC=1,5,2,2,2,0,0,0\r]
Sep 30 00:42:20.61: [ 3273]: --> [2:OK]
Sep 30 00:42:20.61: [ 3273]: <-- [31:AT+FLID="Arlington Fax Forwar"\r]
Sep 30 00:42:20.88: [ 3273]: --> [2:OK]
Sep 30 00:42:20.88: [ 3273]: DIAL 2022250566
Sep 30 00:42:20.88: [ 3273]: <-- [16:ATDT2022250566@\r]
Sep 30 00:42:35.43: [ 3273]: --> [5:+FCON]
Sep 30 00:42:39.12: [ 3273]: --> [141:+FNSF:00 00 51 00 00 A0 B1 C2 F2 72 E2 74 
04 B2 82 2A CA AA 92 04
 04 04 04 04 04 04 04 04 04 04 04 FF FF FF FF FF FF FF FF 41 4E FF FF 00 00 ]
Sep 30 00:42:39.12: [ 3273]: REMOTE NSF "00 00 51 00 00 A0 B1 C2 F2 72 E2 74 04 
B2 82 2A CA AA 92 04 04
 04 04 04 04 04 04 04 04 04 04 FF FF FF FF FF FF FF FF 41 4E FF FF 00 00"
Sep 30 00:42:39.12: [ 3273]: --> [28:+FCSI:"                    "]
Sep 30 00:42:39.12: [ 3273]: REMOTE CSI ""
Sep 30 00:42:39.12: [ 3273]: --> [21:+FDIS:1,5,0,2,1,2,0,1]
Sep 30 00:42:39.12: [ 3273]: --> [2:OK]
Sep 30 00:42:39.12: [ 3273]: REMOTE best rate 14400 bit/s
Sep 30 00:42:39.12: [ 3273]: REMOTE max page width 1728 pixels in 215 mm
Sep 30 00:42:39.12: [ 3273]: REMOTE max unlimited page length
Sep 30 00:42:39.12: [ 3273]: REMOTE best vres 7.7 line/mm
Sep 30 00:42:39.12: [ 3273]: REMOTE best format 2-D MR
Sep 30 00:42:39.12: [ 3273]: REMOTE supports T.30 Annex C, half duplex ECM
Sep 30 00:42:39.12: [ 3273]: REMOTE best 5 ms/scanline
Sep 30 00:42:39.12: [ 3273]: USE 14400 bit/s
Sep 30 00:42:39.12: [ 3273]: USE 5 ms/scanline
Sep 30 00:42:39.12: [ 3273]: SEND file "docq/doc19613.ps;70"
Sep 30 00:42:39.12: [ 3273]: USE page width 1728 pixels in 215 mm
Sep 30 00:42:39.12: [ 3273]: USE unlimited page length
Sep 30 00:42:39.12: [ 3273]: USE 3.85 line/mm
Sep 30 00:42:39.12: [ 3273]: USE 1-D MR
Sep 30 00:42:39.12: [ 3273]: <-- [24:AT+FDIS=0,5,0,2,0,0,0,1\r]
Sep 30 00:42:39.39: [ 3273]: --> [2:OK] 
Sep 30 00:42:39.39: [ 3273]: <-- [7:AT+FDT\r]
Sep 30 00:42:46.75: [ 3273]: --> [21:+FDCS:0,5,0,2,0,0,0,1]
Sep 30 00:42:46.75: [ 3273]: --> [7:CONNECT]
Sep 30 00:42:46.75: [ 3273]: SEND wait for XON
Sep 30 00:42:46.75: [ 3273]: --> [1:^Q]
Sep 30 00:42:46.75: [ 3273]: SEND begin page
Sep 30 00:42:46.75: [ 3273]: <-- data [1028]
Sep 30 00:42:46.75: [ 3273]: <-- data [1026] 
Sep 30 00:42:46.75: [ 3273]: <-- data [1028]
Sep 30 00:42:47.20: [ 3273]: <-- data [1028]
Sep 30 00:42:47.20: [ 3273]: <-- data [1024]
Sep 30 00:42:47.73: [ 3273]: <-- data [1025] 
Sep 30 00:42:47.73: [ 3273]: <-- data [1026]
Sep 30 00:42:50.37: [ 3273]: <-- data [1028]
Sep 30 00:42:50.38: [ 3273]: <-- data [1025]  
Sep 30 00:42:52.91: [ 3273]: <-- data [1024]
Sep 30 00:42:52.91: [ 3273]: <-- data [1024]
Sep 30 00:42:54.42: [ 3273]: <-- data [1025]
Sep 30 00:42:54.42: [ 3273]: <-- data [1025] 
Sep 30 00:42:55.85: [ 3273]: <-- data [1027]
Sep 30 00:42:55.85: [ 3273]: <-- data [1024]
Sep 30 00:42:57.38: [ 3273]: <-- data [1027]
Sep 30 00:42:57.38: [ 3273]: <-- data [1025] 
Sep 30 00:42:58.65: [ 3273]: <-- data [1029]
Sep 30 00:42:58.65: [ 3273]: <-- data [1026]
Sep 30 00:42:59.99: [ 3273]: <-- data [497]
Sep 30 00:42:59.99: [ 3273]: SENT 19952 bytes of data
Sep 30 00:42:59.99: [ 3273]: <-- data [2]
Sep 30 00:42:59.99: [ 3273]: SEND end page  
Sep 30 00:43:03.14: [ 3273]: --> [34:^S^Q^S^Q^S^Q^S^Q^S^Q^S^Q^S^Q^S^Q^S^Q^S^Q^S^
Q^S^Q^S^Q^S^Q^S^Q^S^Q^S
^Q] 
Sep 30 00:43:03.14: [ 3273]: --> [2:OK]
Sep 30 00:43:03.14: [ 3273]: SEND send EOP (no more pages or documents)
Sep 30 00:43:03.14: [ 3273]: <-- [9:AT+FET=2\r]
Sep 30 00:43:08.71: [ 3273]: --> [7:+FPTS:1]
Sep 30 00:43:08.71: [ 3273]: SEND recv MCF (message confirmation)
Sep 30 00:43:08.71: [ 3273]: SEND FAX (00033009): FROM patrick@cre8tivegroup.com
 TO 202-225-0566 (page
1 of 1 sent in 0:29)
Sep 30 00:43:08.71: [ 3273]: SEND FAX (00033009): FROM patrick@cre8tivegroup.com
 TO 202-225-0566 (docq/
doc19613.ps;70 sent in 0:29)
Sep 30 00:43:08.73: [ 3273]: <-- [5:ATH0\r]
Sep 30 00:43:10.18: [ 3273]: --> [7:+FHNG:0]
Sep 30 00:43:10.18: [ 3273]: REMOTE HANGUP: Normal and proper end of connection 
(code 0)
Sep 30 00:43:10.81: [ 3273]: --> [2:OK]
Sep 30 00:43:10.81: [ 3273]: SESSION END

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 16:28:36 -0500 (CDT)
From: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
To: Patrick Gardella <patrick@cre8tivegroup.com>
cc: flexfax@sgi.com, Larry Pyeatt <pyeatt@cs.colostate.edu>
Subject: RE: flexfax: MultiTech MT2834ZDXV recommendation?
Sender: owner-flexfax@celestial.com

Thanks very much for the information,

As there has been somebody else saying that they are having problems with
this same modem, would it be possible for you to let us know the firmware
release and revision date? Hylafax normally prints this out in one of the
log files, during modem reset.

Out of curiosisty what is the failure rate for all causes, except wrong
number or no carrier that you have with these modems?

With USR sportster I am running at around 8-10% and it is unacceptable

Thanks,

Julio

On Wed, 22 Oct 1997, Patrick Gardella wrote:

> I do send 2000+ a day.  Our record day is 4041 in 24 hours on 7 modems.  All the
> faxes are 1 to 2 pages, no cover sheets and all are US (local) numbers.
> 
> Here is my config file for one of the modems:  (The FAXnumber and LocalId are 
> commented out :)
> 
> 
> Patrick
> _________________
> 
> CountryCode:            1
> AreaCode:               703
> FAXNumber:              +xxxxx
> LongDistancePrefix:     1
> InternationalPrefix:    011
> DialStringRules:        "etc/dialrules"
> ServerTracing:          1
> SessionTracing:         11
> RecvFileMode:           0600
> LogFileMode:            0600
> DeviceMode:             0660
> RingsBeforeAnswer:      0 
> SpeakerVolume:          off
> GettyArgs:              "std.%s"
> LocalIdentifier:        "xxxx"
> TagLineFont:            etc/lutRS18.pcf
> TagLineFormat:          "From %%l|%c|Page %%p of %%t"
> MaxRecvPages:           25
> # 
> # 
> # Modem-related stuff: should reflect modem command interface
> # and hardware connection/cabling (e.g. flow control).
> #
> ModemType:              Class2          # use class 2 interface
> ModemRate:              38400           # lock rate for DCE-DTE communication
> ModemFlowControl:       rtscts          # default
> #
> ModemHardFlowCmd:       AT&E4           # hardware flow control
> ModemSoftFlowCmd:       AT&E5           # software flow control
> ModemSetupDTRCmd:       AT&D3           # setup so DTR drop resets modem
> ModemSetupDCDCmd:       AT&C1           # setup so DCD reflects carrier (or not)
> ModemSetupAACmd:        AT+FAA=1        # enable adaptive-answer in class 2
> #
> # NB: some models get confused by the @
> #
> ModemDialCmd:           ATDT%s@         # T for tone dialing, @ for silence
> #
> Class2RecvDataTrigger:  "\022"          # character sent to modem to start recv
> 
> 
> On 22-Oct-97 Prof. Julio C. Spinelli wrote:
> >Dear Patrick,
> >
> >Could you please provide me with the config file that you are using. I
> >would like to try the Multitech modem. Do you really send 2000 faxes per
> >day average? Whow! Are they all to the USA or international?
> >
> >Thanks very much for this help,
> >
> >Modem vendors will learn to market products that really work thaks to the
> >exchange of info that you are doing. Thanks again,
> >
> >Julio
> >
> >On Wed, 22 Oct 1997, Patrick Gardella wrote:
> >
> >> I am using 8 MT2834ZDX modems are they have not given me a problem yet!  What
> >> is the "v" mean?  It this the 28.8 -33.6 update?
> >> 
> >> Been running 24x7 for 3 months at an average of 2000 faxes outgoing. (I don't
> >> allow any incoming.)
> >> 
> >> Patrick
> >> 
> >> On 21-Oct-97 Larry Pyeatt wrote:
> >> >Hello,
> >> >  I have given up on my USR Sportster and am looking for a modem that
> >> >works.
> >> >Has anyone used the MT2834ZDXV with hylafax?  If so, does it work well?
> >> >
> >> >Thanks.
> >> >
> >> >--
> >> >Larry D. Pyeatt                     All standard disclaimers apply.
> >> >pyeatt@cs.colostate.edu             Void where prohibited.
> >> >http://www.cs.colostate.edu/~pyeatt
> >> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
to: flexfax@sgi.com
Subject: Re: flexfax: MultiTech MT2834ZDXV recommendation?
Date: Wed, 22 Oct 1997 17:35:27 -0400
From: Steve Moulton <moulton@snmp.com>
Sender: owner-flexfax@celestial.com


Larry Pyeatt <pyeatt@cs.colostate.edu> wrote:

>  I have given up on my USR Sportster and am looking for a modem that
>works.
>Has anyone used the MT2834ZDXV with hylafax?  If so, does it work well?

I have been using two MODEM MULTI-TECH SYSTEMS MT2834ZDX/0314C
modems for nearly a year serving all of our incoming faxes, after 
giving the USR Sportster modem to someone who had been bugging me.
The USR is just not reliable enough for standalone/server operations, though
as a home modem it seems to be fine.

The 2834s occasionally have a fax come in that looks like the entire
page has been squished into 1/2 inch or so.  Usually repeats from a given
site two or three times, then the problem goes away, so I can't really
blame the modem.

I have been meaning to do outgoing faxes using these, but with three
different document processors being used in a fairly small office
on several different architectures, it has been too much of a 
hassle to interface.


---
Steve Moulton        SNMP Research, Inc            voice: +1 423 573 1434
Software Engineer    3001 Kimberlin Heights Rd.    fax: +1 423 573 9197
moulton@snmp.com     Knoxville, TN 37920-9716      http://www.int.snmp.com
personal home page:  http://www.cs.utk.edu/~moulton

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 16:56:48 -0500 (CDT)
From: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
To: David Birnbaum <davidb@chelsea.net>
cc: Patrick Gardella <patrick@cre8tivegroup.com>, flexfax@sgi.com,
        Larry Pyeatt <pyeatt@cs.colostate.edu>
Subject: RE: flexfax: MultiTech MT2834ZDXV recommendation?
Sender: owner-flexfax@celestial.com

What I would do:
0) Check with manufacturer to see if this is a know problem and try to 
get a later firmware release, revision.
1) Change the modem and use an new one same model
2) If problem persists, change FLID to a numeric string and check again
3) If the problems still persists commit Hara_Kiri.....

Please keep me posted on this subject, I am buying this modem for my
system.

Julio

On Wed, 22 Oct 1997, David Birnbaum wrote:

> I just checked my config file, and it is identical.  Here is a sample
> failure trace:
> 
> Oct 02 19:47:36.55: [ 8646]: SESSION BEGIN 00001980 12126331423
> Oct 02 19:47:36.55: [ 8646]: SEND FAX: JOB 2665 DEST 6331423 COMMID 00001980
> Oct 02 19:47:36.55: [ 8646]: DELAY 2600 ms
> Oct 02 19:47:39.15: [ 8646]: <-- [15:ATE0V1Q0S0=0H0\r]
> Oct 02 19:47:39.40: [ 8646]: --> [14:ATE0V1Q0S0=0H0]
> Oct 02 19:47:39.51: [ 8646]: --> [2:OK]
> Oct 02 19:47:39.51: [ 8646]: <-- [21:ATS8=2S7=60&E4&D3&C1\r]
> Oct 02 19:47:39.83: [ 8646]: --> [2:OK]
> Oct 02 19:47:39.83: [ 8646]: <-- [12:AT+FCLASS=2\r]
> Oct 02 19:47:40.15: [ 8646]: --> [2:OK]
> Oct 02 19:47:40.15: [ 8646]: <-- [10:AT+FTBC=0\r]
> Oct 02 19:47:40.47: [ 8646]: --> [2:OK]
> Oct 02 19:47:40.47: [ 8646]: <-- [10:AT+FBOR=0\r]
> Oct 02 19:47:40.79: [ 8646]: --> [2:OK]
> Oct 02 19:47:40.79: [ 8646]: <-- [13:AT+FPHCTO=30\r]
> Oct 02 19:47:41.10: [ 8646]: --> [2:OK]
> Oct 02 19:47:41.11: [ 8646]: <-- [24:AT+FDCC=1,5,2,2,0,0,0,0\r]
> Oct 02 19:47:41.43: [ 8646]: --> [2:OK]
> Oct 02 19:47:41.43: [ 8646]: <-- [5:ATM0\r]
> Oct 02 19:47:41.75: [ 8646]: --> [2:OK]
> Oct 02 19:47:41.82: [ 8646]: <-- [12:AT+FCLASS=2\r]
> Oct 02 19:47:42.23: [ 8646]: --> [2:OK]
> Oct 02 19:47:42.24: [ 8646]: <-- [10:AT+FTBC=0\r]
> Oct 02 19:47:42.64: [ 8646]: --> [2:OK]
> Oct 02 19:47:42.65: [ 8646]: <-- [10:AT+FBOR=0\r]
> Oct 02 19:47:43.05: [ 8646]: --> [2:OK]
> Oct 02 19:47:43.06: [ 8646]: <-- [13:AT+FPHCTO=30\r]
> Oct 02 19:47:43.46: [ 8646]: --> [2:OK]
> Oct 02 19:47:43.47: [ 8646]: <-- [24:AT+FDCC=1,5,2,2,0,0,0,0\r]
> Oct 02 19:47:43.88: [ 8646]: --> [2:OK]
> Oct 02 19:47:43.89: [ 8646]: <-- [27:AT+FLID="CHELSEA NETWORKS"\r]
> Oct 02 19:48:13.90: [ 8646]: MODEM <Timeout>
> Oct 02 19:48:13.90: [ 8646]: SEND FAILED: Unable to configure modem for fax use
> Oct 02 19:48:13.91: [ 8646]: SESSION END
> 
> Note the timeout.  I know there are some issues with setting the ID to a
> non-numeric parameter, but the fact that it works 97 times out of a
> hundred is what is so strange about it. 
> 
> USR users - does anybody actually use the USR chassis for faxing?  I have
> a MP 8 that I was going to use, but all of the notes from people about USR
> problems has me a bit nervous.
> 
> Incidently, I sent out 600-700 faxes a night using only two modems right
> now....
> 
> David.
> 
> On Wed, 22 Oct 1997, Patrick Gardella wrote:
> 
> > I do send 2000+ a day.  Our record day is 4041 in 24 hours on 7 modems.  All the
> > faxes are 1 to 2 pages, no cover sheets and all are US (local) numbers.
> > 
> > Here is my config file for one of the modems:  (The FAXnumber and LocalId are 
> > commented out :)
> > 
> > 
> > Patrick
> > _________________
> > 
> > CountryCode:            1
> > AreaCode:               703
> > FAXNumber:              +xxxxx
> > LongDistancePrefix:     1
> > InternationalPrefix:    011
> > DialStringRules:        "etc/dialrules"
> > ServerTracing:          1
> > SessionTracing:         11
> > RecvFileMode:           0600
> > LogFileMode:            0600
> > DeviceMode:             0660
> > RingsBeforeAnswer:      0 
> > SpeakerVolume:          off
> > GettyArgs:              "std.%s"
> > LocalIdentifier:        "xxxx"
> > TagLineFont:            etc/lutRS18.pcf
> > TagLineFormat:          "From %%l|%c|Page %%p of %%t"
> > MaxRecvPages:           25
> > # 
> > # 
> > # Modem-related stuff: should reflect modem command interface
> > # and hardware connection/cabling (e.g. flow control).
> > #
> > ModemType:              Class2          # use class 2 interface
> > ModemRate:              38400           # lock rate for DCE-DTE communication
> > ModemFlowControl:       rtscts          # default
> > #
> > ModemHardFlowCmd:       AT&E4           # hardware flow control
> > ModemSoftFlowCmd:       AT&E5           # software flow control
> > ModemSetupDTRCmd:       AT&D3           # setup so DTR drop resets modem
> > ModemSetupDCDCmd:       AT&C1           # setup so DCD reflects carrier (or not)
> > ModemSetupAACmd:        AT+FAA=1        # enable adaptive-answer in class 2
> > #
> > # NB: some models get confused by the @
> > #
> > ModemDialCmd:           ATDT%s@         # T for tone dialing, @ for silence
> > #
> > Class2RecvDataTrigger:  "\022"          # character sent to modem to start recv
> > 
> > 
> > On 22-Oct-97 Prof. Julio C. Spinelli wrote:
> > >Dear Patrick,
> > >
> > >Could you please provide me with the config file that you are using. I
> > >would like to try the Multitech modem. Do you really send 2000 faxes per
> > >day average? Whow! Are they all to the USA or international?
> > >
> > >Thanks very much for this help,
> > >
> > >Modem vendors will learn to market products that really work thaks to the
> > >exchange of info that you are doing. Thanks again,
> > >
> > >Julio
> > >
> > >On Wed, 22 Oct 1997, Patrick Gardella wrote:
> > >
> > >> I am using 8 MT2834ZDX modems are they have not given me a problem yet!  What
> > >> is the "v" mean?  It this the 28.8 -33.6 update?
> > >> 
> > >> Been running 24x7 for 3 months at an average of 2000 faxes outgoing. (I don't
> > >> allow any incoming.)
> > >> 
> > >> Patrick
> > >> 
> > >> On 21-Oct-97 Larry Pyeatt wrote:
> > >> >Hello,
> > >> >  I have given up on my USR Sportster and am looking for a modem that
> > >> >works.
> > >> >Has anyone used the MT2834ZDXV with hylafax?  If so, does it work well?
> > >> >
> > >> >Thanks.
> > >> >
> > >> >--
> > >> >Larry D. Pyeatt                     All standard disclaimers apply.
> > >> >pyeatt@cs.colostate.edu             Void where prohibited.
> > >> >http://www.cs.colostate.edu/~pyeatt
> > >> 
> > 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 08:48:18 -0400 (EDT)
Organization: The Creative Group
From: Patrick Gardella <patrick@cre8tivegroup.com>
To: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
Subject: RE: flexfax: MultiTech MT2834ZDXV recommendation?
Cc: Larry Pyeatt <pyeatt@cs.colostate.edu>, flexfax@sgi.com
Sender: owner-flexfax@celestial.com

I've enabled system tracing for this and will let you know.

My fax report from yesterday was:
Sender                    Pages     Time Pg/min Errs TypRate TypData
patrick@cre8tivegroup.com   416  3:59:02    1.7   25   14400  1-D MR
--------------------------------------------------------------------
Total                       416  3:59:02    1.7   25

Facsimile transmitted since last week:

Sender                    Pages     Time Pg/min Errs TypRate TypData
patrick@cre8tivegroup.com 10039 94:00:25    1.8  532   14400  1-D MR
--------------------------------------------------------------------
Total                     10039 94:00:25    1.8  532

So we are running at about 5% error rate, with most of those wrong numbers/did
not answer.


Of those from the 25, we had 7 Unspecified Phase B errors:
Oct 20 18:18:05.43: [10605]: REMOTE HANGUP: Unspecified Transmit Phase B error
(code 20)
which is about a 1.7% error rate.  And I am pretty sure most of these
were remote hangups.

Patrick


On 22-Oct-97 Prof. Julio C. Spinelli wrote:
>Thanks very much for the information,
>
>As there has been somebody else saying that they are having problems with
>this same modem, would it be possible for you to let us know the firmware
>release and revision date? Hylafax normally prints this out in one of the
>log files, during modem reset.
>
>Out of curiosisty what is the failure rate for all causes, except wrong
>number or no carrier that you have with these modems?
>
>With USR sportster I am running at around 8-10% and it is unacceptable
>
>Thanks,
>
>Julio
>
>On Wed, 22 Oct 1997, Patrick Gardella wrote:
>
>> I do send 2000+ a day.  Our record day is 4041 in 24 hours on 7 modems.  All 
the
>> faxes are 1 to 2 pages, no cover sheets and all are US (local) numbers.
>> 
>> Here is my config file for one of the modems:  (The FAXnumber and LocalId are
 
>> commented out :)
>> 
>> 
>> Patrick
>> _________________
>> 
>> CountryCode:            1
>> AreaCode:               703
>> FAXNumber:              +xxxxx
>> LongDistancePrefix:     1
>> InternationalPrefix:    011
>> DialStringRules:        "etc/dialrules"
>> ServerTracing:          1
>> SessionTracing:         11
>> RecvFileMode:           0600
>> LogFileMode:            0600
>> DeviceMode:             0660
>> RingsBeforeAnswer:      0 
>> SpeakerVolume:          off
>> GettyArgs:              "std.%s"
>> LocalIdentifier:        "xxxx"
>> TagLineFont:            etc/lutRS18.pcf
>> TagLineFormat:          "From %%l|%c|Page %%p of %%t"
>> MaxRecvPages:           25
>> # 
>> # 
>> # Modem-related stuff: should reflect modem command interface
>> # and hardware connection/cabling (e.g. flow control).
>> #
>> ModemType:              Class2          # use class 2 interface
>> ModemRate:              38400           # lock rate for DCE-DTE communication
>> ModemFlowControl:       rtscts          # default
>> #
>> ModemHardFlowCmd:       AT&E4           # hardware flow control
>> ModemSoftFlowCmd:       AT&E5           # software flow control
>> ModemSetupDTRCmd:       AT&D3           # setup so DTR drop resets modem
>> ModemSetupDCDCmd:       AT&C1           # setup so DCD reflects carrier (or n
ot)
>> ModemSetupAACmd:        AT+FAA=1        # enable adaptive-answer in class 2
>> #
>> # NB: some models get confused by the @
>> #
>> ModemDialCmd:           ATDT%s@         # T for tone dialing, @ for silence
>> #
>> Class2RecvDataTrigger:  "\022"          # character sent to modem to start re
cv
>> 
>> 
>> On 22-Oct-97 Prof. Julio C. Spinelli wrote:
>> >Dear Patrick,
>> >
>> >Could you please provide me with the config file that you are using. I
>> >would like to try the Multitech modem. Do you really send 2000 faxes per
>> >day average? Whow! Are they all to the USA or international?
>> >
>> >Thanks very much for this help,
>> >
>> >Modem vendors will learn to market products that really work thaks to the
>> >exchange of info that you are doing. Thanks again,
>> >
>> >Julio
>> >
>> >On Wed, 22 Oct 1997, Patrick Gardella wrote:
>> >
>> >> I am using 8 MT2834ZDX modems are they have not given me a problem yet!  W
hat
>> >> is the "v" mean?  It this the 28.8 -33.6 update?
>> >> 
>> >> Been running 24x7 for 3 months at an average of 2000 faxes outgoing. (I do
n't
>> >> allow any incoming.)
>> >> 
>> >> Patrick
>> >> 
>> >> On 21-Oct-97 Larry Pyeatt wrote:
>> >> >Hello,
>> >> >  I have given up on my USR Sportster and am looking for a modem that
>> >> >works.
>> >> >Has anyone used the MT2834ZDXV with hylafax?  If so, does it work well?
>> >> >
>> >> >Thanks.
>> >> >
>> >> >--
>> >> >Larry D. Pyeatt                     All standard disclaimers apply.
>> >> >pyeatt@cs.colostate.edu             Void where prohibited.
>> >> >http://www.cs.colostate.edu/~pyeatt
>> >> 
>> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 10:13:36 -0400 (EDT)
Organization: The Creative Group
From: Patrick Gardella <patrick@cre8tivegroup.com>
To: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
Subject: RE: flexfax: MultiTech MT2834ZDXV recommendation?
Cc: Larry Pyeatt <pyeatt@cs.colostate.edu>, flexfax@sgi.com
Sender: owner-flexfax@celestial.com

Here it is:

at+fmfr?
+FMFR=Multi-Tech Systems

OK
at+FMDL?
+FMDL=MT2834ZDX

OK
ati1
0316A 

NORTH AMERICAN 

OK
ati4
02-26-1997 14:57


Note for those with Multitechs, ATI1 gives you firmare release and ATI4 gives
release date.

I had the devil of a time getting ServerTracing to work with an setting of 13.  
I logged it to local5 and had /etc/sysconf.conf logging to /usr/fax/etc/syslog
which is where I have HylaFAX installed.  Never got a single message in either
that or /var/log/messages.

I restarted faxq after modifiying etc/config to reflect the above, and
syslogd take the new syslog.conf.

Any suggestions on how to get SystemTracing to work?

Patrick Gardella

On 22-Oct-97 Prof. Julio C. Spinelli wrote:
>Thanks very much for the information,
>
>As there has been somebody else saying that they are having problems with
>this same modem, would it be possible for you to let us know the firmware
>release and revision date? Hylafax normally prints this out in one of the
>log files, during modem reset.
>
>Out of curiosisty what is the failure rate for all causes, except wrong
>number or no carrier that you have with these modems?
>
>With USR sportster I am running at around 8-10% and it is unacceptable
>
>Thanks,
>
>Julio
>
>On Wed, 22 Oct 1997, Patrick Gardella wrote:
>
>> I do send 2000+ a day.  Our record day is 4041 in 24 hours on 7 modems.  All 
the
>> faxes are 1 to 2 pages, no cover sheets and all are US (local) numbers.
>> 
>> Here is my config file for one of the modems:  (The FAXnumber and LocalId are
 
>> commented out :)
>> 
>> 
>> Patrick
>> _________________
>> 
>> CountryCode:            1
>> AreaCode:               703
>> FAXNumber:              +xxxxx
>> LongDistancePrefix:     1
>> InternationalPrefix:    011
>> DialStringRules:        "etc/dialrules"
>> ServerTracing:          1
>> SessionTracing:         11
>> RecvFileMode:           0600
>> LogFileMode:            0600
>> DeviceMode:             0660
>> RingsBeforeAnswer:      0 
>> SpeakerVolume:          off
>> GettyArgs:              "std.%s"
>> LocalIdentifier:        "xxxx"
>> TagLineFont:            etc/lutRS18.pcf
>> TagLineFormat:          "From %%l|%c|Page %%p of %%t"
>> MaxRecvPages:           25
>> # 
>> # 
>> # Modem-related stuff: should reflect modem command interface
>> # and hardware connection/cabling (e.g. flow control).
>> #
>> ModemType:              Class2          # use class 2 interface
>> ModemRate:              38400           # lock rate for DCE-DTE communication
>> ModemFlowControl:       rtscts          # default
>> #
>> ModemHardFlowCmd:       AT&E4           # hardware flow control
>> ModemSoftFlowCmd:       AT&E5           # software flow control
>> ModemSetupDTRCmd:       AT&D3           # setup so DTR drop resets modem
>> ModemSetupDCDCmd:       AT&C1           # setup so DCD reflects carrier (or n
ot)
>> ModemSetupAACmd:        AT+FAA=1        # enable adaptive-answer in class 2
>> #
>> # NB: some models get confused by the @
>> #
>> ModemDialCmd:           ATDT%s@         # T for tone dialing, @ for silence
>> #
>> Class2RecvDataTrigger:  "\022"          # character sent to modem to start re
cv
>> 
>> 
>> On 22-Oct-97 Prof. Julio C. Spinelli wrote:
>> >Dear Patrick,
>> >
>> >Could you please provide me with the config file that you are using. I
>> >would like to try the Multitech modem. Do you really send 2000 faxes per
>> >day average? Whow! Are they all to the USA or international?
>> >
>> >Thanks very much for this help,
>> >
>> >Modem vendors will learn to market products that really work thaks to the
>> >exchange of info that you are doing. Thanks again,
>> >
>> >Julio
>> >
>> >On Wed, 22 Oct 1997, Patrick Gardella wrote:
>> >
>> >> I am using 8 MT2834ZDX modems are they have not given me a problem yet!  W
hat
>> >> is the "v" mean?  It this the 28.8 -33.6 update?
>> >> 
>> >> Been running 24x7 for 3 months at an average of 2000 faxes outgoing. (I do
n't
>> >> allow any incoming.)
>> >> 
>> >> Patrick
>> >> 
>> >> On 21-Oct-97 Larry Pyeatt wrote:
>> >> >Hello,
>> >> >  I have given up on my USR Sportster and am looking for a modem that
>> >> >works.
>> >> >Has anyone used the MT2834ZDXV with hylafax?  If so, does it work well?
>> >> >
>> >> >Thanks.
>> >> >
>> >> >--
>> >> >Larry D. Pyeatt                     All standard disclaimers apply.
>> >> >pyeatt@cs.colostate.edu             Void where prohibited.
>> >> >http://www.cs.colostate.edu/~pyeatt
>> >> 
>> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Leslie Mikesell <les@Mcs.Net>
Subject: Re: flexfax: MultiTech MT2834ZDXV recommendation?
To: patrick@cre8tivegroup.com (Patrick Gardella)
Date: Sun, 2 Nov 1997 21:14:36 -0600 (CST)
Cc: spine001@gold.tc.umn.edu, flexfax@sgi.com, pyeatt@cs.colostate.edu
Sender: owner-flexfax@celestial.com

> I do send 2000+ a day.  Our record day is 4041 in 24 hours on 7 modems.  All the
> faxes are 1 to 2 pages, no cover sheets and all are US (local) numbers.
> 
> Here is my config file for one of the modems:  (The FAXnumber and LocalId are 
> commented out :)
> 
> ModemFlowControl:       rtscts          # default
> #
> ModemHardFlowCmd:       AT&E4           # hardware flow control

One odd thing about the Multitech modems is the the AT&En values
are not mutually exclusive like most AT commands.  AT&E4 sets cts
flow control from modem to the computer.  However you also need
AT&E13 (pacing on) for flow control the other direction from
the computer to the modem via rts if the computer can't keep up
with inbound data.  You can issue this command as AT&E4&E13 and
the modem will work with hardware flow control for both fax and
data.

  Les Mikesell
    les@mcs.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 10 Nov 1997 16:15:59 +0100
From: Sinisha Milivojevich <sinisa@coresinc.com>
Organization: CORES
To: flexfax@sgi.com
Subject: flexfax: Intel SatisFAXtion /400
Sender: owner-flexfax@celestial.com

Two questions from future user:


1. Has anybody been working on the driver for the above board (NOT /400e
!!!!!!!).


2. Does anybody has experience with receiving faxes from fax machines on
the noisy and lousy lines, like the ones we have here in Serbia
????????? Is there anywhere in HylaFax reporting on the reason of
rejection of bad faxes ?????????????


-- 

Sinisa

_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
_/ --- ___            \/          - Sinisa Milivojevic, Consultant  _/
_/ -- /     o   __  o __   ___    - CORES , 11000 Beograd       8-) _/
_/ -  ---  /  /  / / /_   /__/   -- Marsala Tolbuhina 31, SERBIA    _/
_/   ___/ /  /  / / __/  /  /  ---- tel:+381 11 4443331,fax: 433001 _/
_/                            ----- http://www.coresinc.com/~sinisa _/
_/  To defy, to conquer. For us! -- mailto:sinisa@coresinc.com  ;o) _/
_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
Attachment Converted: "C:\PROGRAM FILES\EUDORA\Attach\vcard3.vcf"

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Reply-To: "John Hamill" <jh@lan1.com.au>
From: "John Hamill" <jh@lan1.com.au>
To: "Sinisha Milivojevich" <sinisa@coresinc.com>, <flexfax@sgi.com>
Subject: Re: flexfax: Intel SatisFAXtion /400
Date: Tue, 11 Nov 1997 07:39:40 +1100
Sender: owner-flexfax@celestial.com

There is a class 1, class 2 firmware available for the 400 at
www.wildcard.com. This replaces the proprietary cas interface with standard
class 1 or class 2. However I think you have to boot DOS first, download the
firmware and launch Unix :-)

-----Original Message-----


>Two questions from future user:
>
>
>1. Has anybody been working on the driver for the above board (NOT /400e
>!!!!!!!).
>

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 29 Oct 1997 04:19:14 -0500
From: Ancil Bethelmy <cowpoke@wave.cape.net>
To: HylaFAX Mailing List <flexfax@sgi.com>, Tiff Mailing List <tiff@sgi.com>
Subject: flexfax: Need info on font installation
Sender: owner-flexfax@celestial.com

I have completed the compilation and installation of most of the support
files for setting up HylaFAX. I am now uncertain about what steps to
take to install the fonts for ghostscript.

The list of font files include the 3.33, 4.0 and std fonts tar files. I
have not seen any make files for doing the font file installation.

Any help would be greatly appreciated.

TIA,

--
Ancil Bethelmy
cowpoke@cape.net

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Trevor Blackwell <tlb@eecs.harvard.edu>
X-Authentication-Warning: chardonnay.eecs.harvard.edu: Host localhost didn't use HELO protocol
To: flexfax@sgi.com
Subject: flexfax: Blocking toll numbers
Date: Wed, 29 Oct 97 19:08:45 -0500
X-Mts: smtp
Sender: owner-flexfax@celestial.com


Does anybody know how to block calls to expensive toll numbers? We
provide faxing of orders received from customers web sites, and I want
to protect against someone entering expensive toll numbers.

As I understand it, not just 1-900 numbers, but some numbers in 1-819
and other area codes can be arbitrarily expensive. If there were some
finite list of bad numbers, that would be great.

--
Trevor Blackwell         tlb@eecs.harvard.edu          (617) 495-8912
http://www.eecs.harvard.edu/~tlb

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 30 Oct 1997 10:22:38 +0100
From: Massimo Fusaro <max@leo.inet.it>
To: flexfax@sgi.com
Subject: Re: flexfax: Blocking toll numbers
Reply-To: max@inet.it
Sender: owner-flexfax@celestial.com


look at DestControls configuration parameter and the destctrls man page.

On Wed, Oct 29, 1997 at 07:08:45PM -0500, Trevor Blackwell wrote:
> 
> Does anybody know how to block calls to expensive toll numbers? We
> provide faxing of orders received from customers web sites, and I want
> to protect against someone entering expensive toll numbers.
> 
> As I understand it, not just 1-900 numbers, but some numbers in 1-819
> and other area codes can be arbitrarily expensive. If there were some
> finite list of bad numbers, that would be great.
> 
> --
> Trevor Blackwell         tlb@eecs.harvard.edu          (617) 495-8912
> http://www.eecs.harvard.edu/~tlb

-- 
---
Massimo Fusaro		| E-Mail: max@inet.it
INet S.p.A.		| via Caldera, 21 -Palazzo C2- 20153 Milano

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 30 Oct 1997 10:23:37 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Trevor Blackwell <tlb@eecs.harvard.edu>
Cc: flexfax@sgi.com
Subject: Re: flexfax: Blocking toll numbers
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Wed, 29 Oct 1997, Trevor Blackwell wrote:

> 
> Does anybody know how to block calls to expensive toll numbers? We
> provide faxing of orders received from customers web sites, and I want
> to protect against someone entering expensive toll numbers.
> 
> As I understand it, not just 1-900 numbers, but some numbers in 1-819
> and other area codes can be arbitrarily expensive. If there were some
> finite list of bad numbers, that would be great.

Hmmm. One could add some filtering into the etc/dialrules file to turn
such numbers into your local fax number: this would let you track when
people do this....

Depending on your local phone system, it may also be possible to block
at the telephone. Many office systems, for example, require some sort
of permission to make long distance calls.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu


-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNFimcj/+ItycgIJRAQEi1QP9H6Xhb3/FlioIW/1pSxHB6dSLPiCenpwk
iwRqCMantZMg2TKflim3IHiLnhnWWySHjKbuWhLdfCDzoiF5dLOl4GAGZTDPobRs
Wlh9HSFvcAhbKOi4NH4yHl22hxvui22l3ncYmAq6dpCBe+JRM3Lxit05nJkWbj7L
j2zlNlP/KD4=
=9LQR
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: "Alan Sparks" <asparks@nss.harris.com>
To: <flexfax@sgi.com>
Subject: flexfax: Begging suggestions on "document page width" errors
Date: Mon, 27 Oct 1997 15:42:05 -0800
Sender: owner-flexfax@celestial.com

 Think I'm losing my mind here, with these "Client does not support document
page width, max remote page width 1728 pixels, image width 1734 pixels"
messages!

I'm trying to send docs out of MS Word, trying a variety of PostScript
drivers.
I've tried at least the following:
* HP LaserJet 4/4M PS
* AST TurboLazer PS
* Apple Laserwriter

Checking what's running inside of ps2fax, I am running the command (after
variable substitution):

cat docq/doc71.ps.63  |  gs -q -sDEVICE=tiffg3 -dNOPAUSE \
    -dSAFER=true -sPAPERSIZE=a4 -r204x98  \
    "-sOutputFile=docq/doc71.ps;00" -

I've checked script options, which seem to show that the expected dimensions
are indeed 1728x297.

All of which leaves me with a TIFF file, about which tiffinfo says:
    Image Width: 1734 Image Length 1078
    Resolution: 204, 98 pixels/inch
    Bits/Sample: 1
    Software: "Alladin Ghostscript 5.03"
    Group 3 Options: EOL padding (4 = 0x4)
    ... etc ...

How the devil do you get 1728 pixels wide out of this stupid thing?
Thanks in advance for any advice.
-Alan

---------
Alan Sparks, IS Engineering Support                   <asparks@harris.com>
Harris Network Support Systems, Camarillo CA    (805) 389-2430

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 27 Oct 1997 22:04:35 -0800 (PST)
From: Tim Rice <tim@trr.metro.NET>
To: Alan Sparks <asparks@nss.harris.com>
cc: flexfax@sgi.com
Subject: Re: flexfax: Begging suggestions on "document page width" errors
Sender: owner-flexfax@celestial.com


Here a a couple of solutions that came from the list.
I suspect they are in the mail archive.

: Subject: FrameMaker ps file Solution!
: Date: Mon, 14 Apr 1997 18:44:35 JST
: From: "Andrew S. Howell" <andy@tibco.com>
: 
: I finally got FrameMaker PS documents to work with gs 4.03. This is a
: bit of a hack, but it works.
: 
: FrameMaker docs want to set there own size when they print. This
: causes problems because the page comes out 1734 pixles wide, instead
: of 1728. Fixing involves changing ps2fax to use -dFIXEDMEDIA, the tail
: end of which now look like:
: 
: a$CAT $fil | $PS -q \
:     -sDEVICE=$device \
:     -dNOPAUSE \
:     -dSAFER=true \
:     -sPAPERSIZE=$paper \
:     -dFIXEDMEDIA \
:     -r204x$vres \
:     "-sOutputFile=$out" \
:     -
: With this alone though, FrameMaker docs just spit out a message that
: they can't set the page size. I then I added the following to gs_init.ps:
: 
: % Ignore PageSize change requests. This is to make FrameMaker documents 
: % work with Hylafax. ps2fax needs -dFIXEDMEDIA as well. 
: << /Policies << /PageSize 1 >> >> setpagedevice
: 
: Since gs looks for its files in the current directory, I copied
: gs_init.ps to where I have my Hylafax installed. In my case,
: /var/spool/fax. This way my kludge won't interfere with other uses of
: gs. When ps2fax runs, the current directory is /var/spool/fax
: 
: Having done that, FrameMaker 4.0 and 5.0 documents that were formated
: for letter size come out just fine. Horraaaayyyy!
: 
: Andy
: 
: Date: Wed, 13 Aug 1997 12:21:25 -0400
: From: Jon Earle <jearle@hubcc.ca>
: Subject: flexfax: FIX for the PAGEWIDTH (Pixelsize) Problem
: 
: in the PS2FAX file, modify the last portion of the script which tosses the
: file off to Ghostscript as follows:
: 
: Change the first line of this from: 
: 
: 	$CAT $fil | $PS -q \
: 
: to:
: 	$CAT $fil | $SED -e 's/dup \/PageSize \[612 792\] put//g' | $PS -q \
: 
: This removes the pagesize info from the postscript file, and all seems to
: work fine thereafter.
: 
: NOTE:  This works on OUR system which is Linux kernel 2.0.29, HylaFAX
: 4.0pl1, WHFC client .3e.  We take no responsibility if this sends obnoxious
: X-Rated faxes to the White House or other such nonsense.
: 
: Cheers!
: Jon


On Mon, 27 Oct 1997, Alan Sparks wrote:

>  Think I'm losing my mind here, with these "Client does not support document
> page width, max remote page width 1728 pixels, image width 1734 pixels"
> messages!
> 
> I'm trying to send docs out of MS Word, trying a variety of PostScript
> drivers.
> I've tried at least the following:
> * HP LaserJet 4/4M PS
> * AST TurboLazer PS
> * Apple Laserwriter
> 
> Checking what's running inside of ps2fax, I am running the command (after
> variable substitution):
> 
> cat docq/doc71.ps.63  |  gs -q -sDEVICE=tiffg3 -dNOPAUSE \
>     -dSAFER=true -sPAPERSIZE=a4 -r204x98  \
>     "-sOutputFile=docq/doc71.ps;00" -
> 
> I've checked script options, which seem to show that the expected dimensions
> are indeed 1728x297.
> 
> All of which leaves me with a TIFF file, about which tiffinfo says:
>     Image Width: 1734 Image Length 1078
>     Resolution: 204, 98 pixels/inch
>     Bits/Sample: 1
>     Software: "Alladin Ghostscript 5.03"
>     Group 3 Options: EOL padding (4 = 0x4)
>     ... etc ...
> 
> How the devil do you get 1728 pixels wide out of this stupid thing?
> Thanks in advance for any advice.
> -Alan
> 
> ---------
> Alan Sparks, IS Engineering Support                   <asparks@harris.com>
> Harris Network Support Systems, Camarillo CA    (805) 389-2430
> 
> 
> 

--
Tim Rice			Multitalents	(707) 887-1469 (voice)
tim@trr.metro.net

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Begging suggestions on "document page width" errors
To: asparks@nss.harris.com (Alan Sparks)
Date: Tue, 28 Oct 1997 08:02:20 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Alan Sparks wrote:
    
     Think I'm losing my mind here, with these "Client does not support document
    page width, max remote page width 1728 pixels, image width 1734 pixels"
    messages!
    
    I'm trying to send docs out of MS Word, trying a variety of PostScript
    drivers.
    I've tried at least the following:
    * HP LaserJet 4/4M PS
    * AST TurboLazer PS
    * Apple Laserwriter
    
    Checking what's running inside of ps2fax, I am running the command (after
    variable substitution):
    
    cat docq/doc71.ps.63  |  gs -q -sDEVICE=tiffg3 -dNOPAUSE \
        -dSAFER=true -sPAPERSIZE=a4 -r204x98  \
        "-sOutputFile=docq/doc71.ps;00" -
    
    I've checked script options, which seem to show that the expected dimensions
    are indeed 1728x297.
    
    All of which leaves me with a TIFF file, about which tiffinfo says:
        Image Width: 1734 Image Length 1078
        Resolution: 204, 98 pixels/inch
        Bits/Sample: 1
        Software: "Alladin Ghostscript 5.03"
        Group 3 Options: EOL padding (4 = 0x4)
        ... etc ...
    
    How the devil do you get 1728 pixels wide out of this stupid thing?
    Thanks in advance for any advice.

Looks for me like landscape mode in the PostScript file?

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 16 Oct 1997 16:03:31 -0500
From: Lachlan Dunlop <lach@lach.net>
To: flexfax@sgi.com
Subject: flexfax: Hylafax and form printing
Sender: owner-flexfax@celestial.com

Hello,

Hylafax is now up and running and it looks like most of my problems stem
from the use of whfc and US paper sizes.  Once set to A4 everything
worked fine.  THANKS!!!

Now on to the fun stuff......

Is there any solutions in hylafax land for overlay or form printing???
I have a group of plain text reports that I would like to overlay with a
nice form in memory before faxing it.  I do this now with macros under
PCL language.

Any input would be appreciated.

Lach

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 15 Oct 1997 00:37:37 -0500
To: flexfax@sgi.com
From: Michael Castellano <webmaster@gcstation.net>
Subject: flexfax: Need a Little Help Please
Sender: owner-flexfax@celestial.com

I am running Redhat 4.0 just installed newest hylafax.

When I do a faxstat it shows to be running, but when I send a test
and do a faxstat -ad I get the status as "Can't find initialization"

I did post this question a few days back but got no responce.
Can anyone point me in a direction ?

Thanks in advance

Mike
mikec@gcstation.net

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 15 Oct 1997 13:56:29 +0200
From: Claus Svarer <csvarer@pet.rh.dk>
To: flexfax@sgi.com
Subject: flexfax: Problem with faxq - Hylafax v4.0pl1 at HP-UX 10.20
Sender: owner-flexfax@celestial.com

Hi.

I have a question about the compilation and use of Hylafax 4.0pl1 at a
HP900/712 with HP-UX 10.20.  I have compiled the program using the HP
compiler C++ (CC) and ansi-c (cc -Aa). With some smaller changes to the
configuration files the program seems to compile and install without any
problems. Also the configuration of modems and the start of hfaxd deamon
seems seems to be without any problems. The hfaxd response to the net
ports.

Trying to start he faxq daemon I see the problem The program starts
without any complaints, but finish just as fast (the same is for the
faxgetty program). No logging of output is made in /usr/spool/fax/log
or  /usr/adm/syslog/syslog.log. Does anyone have any idea about what the
problem could be?

Thanks for your collaboration
Claus Svarer

---

                               -
 _____________________________o
o__________________________________________
|
L                                           |
| Claus Svarer
\_/                                          |
| Neurobiology Research Unit   | Phone  : +45 3545
6716                    |
| National University Hospital
|                                           |
| Rigshospitalet, build. 9201  | Fax    : +45 3545
6713                    |
| Blegdamsvej 9
|                                           |
| DK-2100 Copenhagen O         | e-mail :
csvarer@pet.rh.dk                |
| Denmark                      |
http://neuro.pet.rh.dk/staff/csvarer.html |
|______________________________|___________________________________________|

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 15 Oct 1997 15:14:06 -0700 (PDT)
From: Michael Oswell <faxmail@page.esgroup.net>
To: flexfax@celestial.com
Subject: flexfax: Phone/Fax combination lines
Sender: owner-flexfax@celestial.com

I'm running hylafax on a RedHat Linux 4.2 (Biltmore) machine (Pentium
133), with a USR 33.6 Sportster modem.  I've been running paging for
several months now without a problem, as well as some ocasional faxes.

Recently, i've run into a problem faxing to fax machines that share phone
lines with a company's phone. (ie. Same phone number for voice and fax.).

It seems as though hylafax is phoneing the number, getting several rings,
then connecting.  But it drops the connection before anything happens.
The log clipping below is from one of these calls, and saysno carrier is
the problem, though we can hear it handshaking.

Anyone have any ideas?
Thanks.

Michael Oswell
ESI Communications Inc.

Oct 14 13:35:05.51: [25326]: --> [10:NO CARRIER]
Oct 14 13:35:05.51: [25326]: SEND FAILED: No carrier detected
Oct 14 13:35:05.51: [25326]: <-- [5:ATH0\r]
Oct 14 13:35:05.74: [25326]: --> [2:OK]
Oct 14 13:35:05.74: [25326]: SESSION END

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Rudi Heitbaum <rudi@darx.com.au>
Subject: flexfax: Resetting the faxmodem after a send?
To: flexfax@sgi.com
Date: Thu, 16 Oct 1997 21:41:50 +1000 (EST)
Sender: owner-flexfax@celestial.com

I have HylaFax 4.0p1 running on Linux 2.0.29 and after a fax has been
sent the faxmodem is left in an inconsistent state. 

The fax is sent at 0:11, the fax line rings at 09:09 and the faxmodem
answers the line, and immediately disconnects. Can someone provide some
light on why the modem is not reseting or hanging up after a send.
Here are the log files.

TIA

Rudi

.......End of sent fax 0:13.......
Oct 16 00:13:06.53: [23542]: <-- [5:ATH0\r]
Oct 16 00:13:08.20: [23542]: --> [8:+FHNG: 0]
Oct 16 00:13:08.20: [23542]: REMOTE HANGUP: Normal and proper end of connection (code 0)
Oct 16 00:13:08.78: [23542]: --> [2:OK]
Oct 16 00:13:08.80: [23542]: SESSION END

.......Fax receive 09:09......
Oct 16 09:09:07.22: [27150]: SESSION BEGIN 00000159 61398221728
Oct 16 09:09:07.22: [27150]: <-- [4:ATA\r]
Oct 16 09:09:16.58: [27150]: --> [8:+FHNG: 2]
Oct 16 09:09:16.75: [27150]: REMOTE HANGUP: Call aborted,  from +FK or <CAN> (code 2)
Oct 16 09:09:16.75: [27150]: ANSWER: Ring detected without successful handshake
Oct 16 09:09:16.75: [27150]: SESSION END

.......Syslog.......
Oct 16 00:11:29 iago HylaFAX[23532]: Filesystem has SysV-style file creation semantics.
Oct 16 00:11:31 iago FaxQueuer[1148]: SUBMIT JOB 77
Oct 16 00:11:45 iago FaxSend[23542]: MODEM ROCKWELL AC/V34/V1.450-V34_DS-0.76
Oct 16 00:11:45 iago FaxSend[23542]: SEND FAX: JOB 77 DEST 03 9870 8928 COMMID 00000158
Oct 16 00:13:08 iago FaxQueuer[1148]: NOTIFY: bin/notify "doneq/q77" "done" "1:28"

Oct 16 09:09:16 iago FaxGetty[27150]: ANSWER: Ring detected without successful handshake
Oct 16 09:09:21 iago FaxGetty[27150]: MODEM ROCKWELL AC/V34/V1.450-V34_DS-0.76

Oct 16 09:10:59 iago FaxGetty[27150]: ANSWER: FAX CONNECTION
Oct 16 09:11:24 iago FaxGetty[27150]: RECV FAX (00000160): from 61 3 95238932, page 1 in 0:23, INF, 3.85 line/mm, 2-D MR
Oct 16 09:11:24 iago FaxGetty[27150]: RECV FAX (00000160): recvq/fax00050.tif from 61 3 95238932, route to <unspecified>, 1 pages in 0:24
Oct 16 09:11:27 iago FaxGetty[27150]: RECV FAX: bin/faxrcvd "recvq/fax00050.tif" "ttyS1" "00000160" ""
Oct 16 09:11:33 iago FaxGetty[27150]: MODEM ROCKWELL AC/V34/V1.450-V34_DS-0.76
.......end Syslog...........

--
Rudi Heitbaum                Ph:  +61-3-9822 1216
Managing Director            Fax: +61-3-9822 1728 
DARX Consulting Pty Ltd      http://www.darx.com.au/
mailto:rudi@darx.com.au      http://www.vsg.edu.au/

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 16 Oct 1997 12:09:49 -0500 (EST)
From: Dave Lewis <dlewis@appn.ci.in.ameritech.com>
To: flexfax@sgi.com
Subject: flexfax: failure to train, request help
Sender: owner-flexfax@celestial.com

Hi, I'm Dave Lewis at Ameritech, in Indianapolis, and we have been 
extensively using Hylafax s/w for several years now.  Our current setup:
binary release v4.0p10, for AIX
AIX version:  4.1.4
our main fax server system has 9 modems, 7 are USR 14400, one is a 
Multi-Tec, and one is a Hayes Accura 28800.

I'm new both to this company (Ameritech) and Hylafax s/w, so please 
forgive me if my problem is an old one that you've seen before.  Overall, 
considering the volume of faxes we send out, Hylafax s/w is working 
great, and I like the 'configurability' and thoroughness of this s/w.
The problem we are having is with a few (probably about 10 out of over 
700 destination fax machines which we send to) destination fax machines, 
which seem to be Pitney Bowes, that do not handshake with the USR 
modems.  We get failure to train after 3 attempts.  However, when I 
manually direct faxes to these problem machines, and specify the Multi 
Tech or Hayes to be used by sendfax, they handshake just fine.  I'm 
reluctant to change configurations in /var/spool/fax/etc/config.tty# for 
the USR modems, since most faxes go-out just fine through these modems.  
What I want to know:  is there a means of configuring some aspect of 
Hylafax s/w to, based on the destination fax machine phone number, to 
direct these faxes to a particular modem on our server?  (i.e., to the 
Multi Tech or Hayes?)  The application developers are reluctant to change 
their code when sendfax is invoked, to specify certain modems (the -h 
option), based on a few problem numbers.  (Or do you know exactly how to 
get the USR to handshake with Pitney Bowes 9230F machines?)  
One thing that I will do is try to setup priorities on the modem pool, so 
that the Hayes and Multi Tech modems will tend to send out most of the 
faxes;  this should at least help.  
Any insight is much appreciated.  Thanks much, Dave Lewis (317)265-4652
  

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 17 Oct 1997 17:13:33 +0200 (MET DST)
From: Magnus Risberg <magnus@risberg.se>
To: flexfax@sgi.com
Subject: flexfax: Two faxcovers per fax
Sender: owner-flexfax@celestial.com


Hi,
I am new to this list and also to the Hylafax software. 

I am running Hylafax V4.0 Patch Level 1 which resides on a 
Sun Sparc with Solaris 2.5.1.
The modem is a USRobotics Courier V.Everything:
   
   Options                HST,V32bis,Terbo,VFC,V34+,x2
   Fax Options            Class 1,Class 2.0
   Clock Freq             20.16Mhz
   Flash ROM              512k
   Ram                    64k/Supervisor date        04/23/97
   DSP date               04/21/97
   Supervisor rev         046-7.1.8
   DSP rev                2.2.5
   
I have problems with the faxcover page.
The software produces two covers per session/fax.

I have also tried other modems without improvement.

What is wrong? Anyone else having this problem/feature?
I can't find any info in the documentation nor in the FAQs.
Any help would be appreciated.

Thanks,

	/Magnus

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Sat, 18 Oct 1997 11:34:10 +0200 (EEST)
From: kees <kees@schoen.nl>
To: Magnus Risberg <magnus@risberg.se>
Cc: flexfax@sgi.com
Subject: Re: flexfax: Two faxcovers per fax
Sender: owner-flexfax@celestial.com


On Fri, 17 Oct 1997, Magnus Risberg wrote:

[snip]

> I have problems with the faxcover page.
> The software produces two covers per session/fax.

Do you mean: 2 the SAME pages or one blank page.
I found that the makefaxcover script generates a 'showpage'
extra at the end.

Kees

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Sun, 19 Oct 1997 00:27:30 +0200 (MET DST)
From: Magnus Risberg <magnus@risberg.se>
To: kees <kees@schoen.nl>
cc: flexfax@sgi.com
Subject: Re: flexfax: Two faxcovers per fax
Sender: owner-flexfax@celestial.com

> 
> [snip]
> 
> > I have problems with the faxcover page.
> > The software produces two covers per session/fax.
> 
> Do you mean: 2 the SAME pages or one blank page.
> I found that the makefaxcover script generates a 'showpage'
> extra at the end.
> 
> Kees

Actually, they look different from each other.
But togheter they have the total amount of information I want.

For instance, the first page doesn't show the subject but it 
shows the from-adress. The second page doesn't show the 
from-field but the subject...

	/Magnus

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 17 Oct 1997 14:28:18 -0700 (PDT)
From: Thomas Unger <unger@raindrop.seaslug.org>
To: flexfax@sgi.com
Subject: flexfax: Adaptive answer.
Sender: owner-flexfax@celestial.com



Hi again.  I'm running HylaFax 4.0 on freebsd with a ZOOM K56.

I've hooked up this modem and have started working with getting it to
work.  In hand tests it seems to do adaptive answer BUT for data mode
I need to have the following occur:

  1)  modem says "RING"
  2)  send command "ATA"
  3)  modem says "DATA"
  4)  send command "ATO"
  5)  modem says "CONNECT xxxx"

The key is to send ATO.  And I can do this by setting
"ModemAnswerDataBeginCmd: ATO".  But it leads to the following:


Oct 17 13:31:18.82: [10132]: SESSION BEGIN 00000138 12063027261
Oct 17 13:31:18.82: [10132]: <-- [4:ATA\r]
Oct 17 13:31:39.20: [10132]: --> [4:DATA]
Oct 17 13:31:39.20: [10132]: <-- [4:ATO\r]
Oct 17 13:31:40.33: [10132]: --> [14:CONNECT 115200]
Oct 17 13:32:10.34: [10132]: MODEM TIMEOUT: reading line from modem
Oct 17 13:32:10.34: [10132]: MODEM <Timeout>
Oct 17 13:32:10.35: [10132]: ANSWER: DATA CONNECTION
Oct 17 13:32:10.35: [10132]: GETTY: START "/usr/libexec/getty std.115200",
pid 10136
Oct 17 13:32:10 raindrop FaxGetty[10132]: ANSWER: DATA CONNECTION
Oct 17 13:32:10 raindrop FaxGetty[10132]: GETTY: START "/usr/libexec/getty
std.1


The modem seems to miss the "CONNECT" message.  Then it times out after
30 seconds and starts getty.  So this works but I would like to eliminate
the delay.  

Suggestions?


                                                      ooooo
 -- Tom Unger                                       oo\    oo
    Wet Weather Consulting                         o   \     o
    (206)325-7403                                 o  B I\V I  o
                                                   o     \   o
                                                    oo    \oo
                                                      ooooo

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
X-Authentication-Warning: tui.pinnacle.co.nz: jonc owned process doing -bs
Date: Mon, 20 Oct 1997 13:18:54 +1300 (NZDT)
From: Jonathan Chen <jonc@pinnacle.co.nz>
To: Thomas Unger <unger@raindrop.seaslug.org>
cc: flexfax@sgi.com
Subject: Re: flexfax: Adaptive answer.
Sender: owner-flexfax@celestial.com

On Fri, 17 Oct 1997, Thomas Unger wrote:

> Hi again.  I'm running HylaFax 4.0 on freebsd with a ZOOM K56.
> 
> I've hooked up this modem and have started working with getting it to
> work.  In hand tests it seems to do adaptive answer BUT for data mode
> I need to have the following occur:
> 
>   1)  modem says "RING"
>   2)  send command "ATA"
>   3)  modem says "DATA"
>   4)  send command "ATO"
>   5)  modem says "CONNECT xxxx"

Try idling in class 0 (ie data) mode instead, and this may eliminate
the manual switching with ATO that appears that you're having to do.
At a guess, you could possibly enable this with

	ModemSetupAACmd:	AT+FAA=1;AT+FCLASS=0

or
	ModemSetupAACmd:	AT+FAA=1+FCLASS=0

(some modems hate the `;')
--
Jonathan Chen                          e-mail : jonc@pinnacle.co.nz
  Pinnacle Software Ltd                Voice  : +64.9.415.4460
  Auckland, New Zealand                Fax    : +64.9.415.4250

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 00:06:46 -0700 (PDT)
From: Thomas Unger <unger@raindrop.seaslug.org>
To: Jonathan Chen <jonc@pinnacle.co.nz>
cc: flexfax@sgi.com
Subject: Re: flexfax: Adaptive answer.
Sender: owner-flexfax@celestial.com


> Try idling in class 0 (ie data) mode instead, and this may eliminate
> the manual switching with ATO that appears that you're having to do.
> At a guess, you could possibly enable this with

Actually that is what I do.  Here's an extract from the log that shows
faxgetty setting up the modem for answer:

Oct 21 23:44:52 raindrop FaxGetty[10178]: MODEM Supports 40 ms/scanline
Oct 21 23:44:52 raindrop FaxGetty[10178]: <-- [9:AT+FRM=?\r]
Oct 21 23:44:52 raindrop FaxGetty[10178]: -->
[41:3,24,48,72,73,74,96,97,98,121,122,145,146]
Oct 21 23:44:52 raindrop FaxGetty[10178]: --> [2:OK]
Oct 21 23:44:52 raindrop FaxGetty[10178]: <-- [12:AT+FCLASS=1\r]
Oct 21 23:44:52 raindrop FaxGetty[10178]: --> [2:OK]
Oct 21 23:44:52 raindrop FaxGetty[10178]: <-- [18:AT+FCLASS=0+FAA=1\r]
Oct 21 23:44:52 raindrop FaxGetty[10178]: --> [2:OK]
Oct 21 23:44:52 raindrop FaxGetty[10178]: MODEM  /
Oct 21 23:44:52 raindrop FaxGetty[10178]: <-- [5:ATM0\r]
Oct 21 23:44:52 raindrop FaxGetty[10178]: --> [2:OK]
Oct 21 23:44:52 raindrop FaxGetty[10178]: MODEM input buffering disabled


                                                      ooooo
 -- Tom Unger                                       oo\    oo
    Wet Weather Consulting                         o   \     o
    (206)325-7403                                 o  B I\V I  o
                                                   o     \   o
                                                    oo    \oo
                                                      ooooo

On Mon, 20 Oct 1997, Jonathan Chen wrote:

> Date: Mon, 20 Oct 1997 13:18:54 +1300 (NZDT)
> From: Jonathan Chen <jonc@pinnacle.co.nz>
> To: Thomas Unger <unger@raindrop.seaslug.org>
> Cc: flexfax@sgi.com
> Subject: Re: flexfax: Adaptive answer.
> 
> On Fri, 17 Oct 1997, Thomas Unger wrote:
> 
> > Hi again.  I'm running HylaFax 4.0 on freebsd with a ZOOM K56.
> > 
> > I've hooked up this modem and have started working with getting it to
> > work.  In hand tests it seems to do adaptive answer BUT for data mode
> > I need to have the following occur:
> > 
> >   1)  modem says "RING"
> >   2)  send command "ATA"
> >   3)  modem says "DATA"
> >   4)  send command "ATO"
> >   5)  modem says "CONNECT xxxx"
> 
> Try idling in class 0 (ie data) mode instead, and this may eliminate
> the manual switching with ATO that appears that you're having to do.
> At a guess, you could possibly enable this with
> 
> 	ModemSetupAACmd:	AT+FAA=1;AT+FCLASS=0
> 
> or
> 	ModemSetupAACmd:	AT+FAA=1+FCLASS=0
> 
> (some modems hate the `;')
> --
> Jonathan Chen                          e-mail : jonc@pinnacle.co.nz
>   Pinnacle Software Ltd                Voice  : +64.9.415.4460
>   Auckland, New Zealand                Fax    : +64.9.415.4250
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Sun, 19 Oct 1997 12:18:44 -0500
From: "Julio C. Spinelli" <spine001@gold.tc.umn.edu>
Organization: University of Minnesota
To: rob@eats.com
CC: flexfax@sgi.com
Subject: flexfax: Would like to write about vgetty and egetty
X-URL: http://www.vix.com/cgi-bin/mlafetch/hylafax/Archives/?Msg=11363&Subj=egetty&SM=10&SD=12&SY=93&EM=10&ED=19&EY=97&SC=t&RV=Yes&IC=Yes&Hits=AAD0qnAAACgIC9BU0yCVq1LLDqu3OCDcRLJBAAAAB3
Sender: owner-flexfax@celestial.com

Rob,

Could I get from you or somewhere else a bore bones example of using egetty to just detect a fax call. I don't mean 
really detecting it, but a shell that does nothing but returns "2" so that faxgetty answers the call as fax.

I have attempted to do so myself, and everything works but after egetty returns and faxgetty starts answering a fax 
call, we get a modem write error, the error is coming from a call to the Sys::write(fd, ... function.

Here is the tracing of these events, If you could help me I would like to clarify this subject for good.

Thanks in advance,

Julio Spinelli

>From the tracer:

Oct 18 16:33:51 puma FaxGetty[21165]: MODEM USROBOTICS SPORTSTER V.34+ FAX /
Oct 18 16:33:51 puma FaxGetty[21165]: <-- [5:ATM1\r]
Oct 18 16:33:51 puma FaxGetty[21165]: --> [2:OK]
Oct 18 16:33:51 puma FaxGetty[21165]: <-- [9:AT+FCR=1\r]
Oct 18 16:33:51 puma FaxGetty[21165]: --> [2:OK]
Oct 18 16:33:51 puma FaxGetty[21165]: <-- [20:AT+FLI="6124845882"\r]
Oct 18 16:33:52 puma FaxGetty[21165]: --> [2:OK]
Oct 18 16:33:52 puma FaxGetty[21165]: <-- [9:AT+FAA=0\r]
Oct 18 16:33:52 puma FaxGetty[21165]: --> [2:OK]
Oct 18 16:33:55 puma FaxGetty[21165]: --> [4:RING]
Oct 18 16:33:56 puma FaxGetty[21165]: EXTERN GETTY: START "/bin/egetty cua2 38400", pid 21171
Oct 18 16:33:56 puma FaxGetty[21165]: EXTERN GETTY: exit status 01000
Oct 18 16:33:56 puma FaxGetty[21165]: ANSWER: FAX CONNECTION
Oct 18 16:33:56 puma FaxGetty[21165]: <-- [5:ATH0\r]
Oct 18 16:33:56 puma FaxGetty[21165]: MODEM WRITE ERROR: errno 5
Oct 18 16:33:56 puma FaxGetty[21165]: <-- [17:ATE0V1Q0X4S0=0H0\r]
Oct 18 16:33:57 puma FaxGetty[21165]: --> [2:OK]
Oct 18 16:33:57 puma FaxGetty[21165]: <-- [32:ATS8=2S7=60&H1&I0&R2S13=1&D2&C1\r]
Oct 18 16:33:57 puma FaxGetty[21165]: --> [2:OK]
Oct 18 16:33:57 puma FaxGetty[21165]: <-- [12:AT+FCLASS=?\r]
Oct 18 16:33:57 puma FaxGetty[21165]: --> [7:0,1,2.0]
Oct 18 16:33:57 puma FaxGetty[21165]: --> [2:OK]
Oct 18 16:33:57 puma FaxGetty[21165]: <-- [14:AT+FCLASS=2.0\r]
Oct 18 16:33:58 puma FaxGetty[21165]: --> [2:OK]
Oct 18 16:33:58 puma FaxGetty[21165]: <-- [8:AT+FMI?\r]
Oct 18 16:33:58 puma FaxGetty[21165]: --> [31:USRobotics Sportster V.34+ FAX ]

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 20 Oct 1997 09:32:39 +0100
To: HylaFAX List <flexfax@sgi.com>
From: Phil Watkinson <pkw@elgro.co.uk>
Subject: flexfax: egetty and vgetty
Cc: Julio C Spinelli <spine001@gold.tc.umn.edu>
Sender: owner-flexfax@celestial.com


On Fri, 17 Oct 1997, Julio C Spinelli wrote :
>It would be great if you could add how to use egetty and give a very 
>simple example that does nothing but recognized the call as a fax 
>call. This will help a lot of people (including me). Same would be 
>great for vgetty.

You have the advantage over me here, since I know nothing about egetty
( I'm only the piano player..). Like what does the 'e' suppose to stand
for?  Q123 of the FAQ doesn't give much detail; Nico Garcia asked a 
couple of weeks ago for the location of the egetty program, with no 
reply; so I guess hard information about it maybe difficult to come by. 

A description of Vgetty and its homepage is mentioned on the contrib.html
page, as part of the section on associated programs to HylaFAX. The
TODO List (todo.html) also acknowledges that integration issues with
vgetty are a high priority - it is required to implement a 'faxback'
facility.

[ A faxback system is where a person calls in, hears a string of menus
  which they navigate with the keypad of their phone. When they decide
  on a document they want, the system faxes it back to the fax number
  of their choice ].

If you are interested in doing some work in this area, you will find 
more information by searching the mailing list archives for emails from
'Ivan Kohler' and 'Bret A. Schuhmacher'.

Cheers,
Phil Watkinson,
Boston, UK.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Comments: Authenticated sender is <spine001@gold.tc.umn.edu>
From: Julio C Spinelli <spine001@gold.tc.umn.edu>
To: Sam Leffler <sam@engr.sgi.com>, Phil Watkinson <pkw@elgro.co.uk>
Date: Mon, 20 Oct 1997 23:34:29 +0000
Subject: flexfax: Re: egetty and vgetty
CC: HylaFAX List <flexfax@sgi.com>
Sender: owner-flexfax@celestial.com

Thanks for the info. egetty is nothing, it does not exist at all. The 
only meaning of the string egetty is that if EGettyArgs is defined as 
a non nul string and AnswerRotary is set to "extern". Hylafax will 
call a program /bin/egetty. If that program does not exist...

It is the user/system developer's responsibility to write that 
program. This program's job is to answer the incoming call, determine 
its type and then exit with a CALLTYPE_XXX (FAX, DATA, DONE, ERROR, 
VOICE) code. Hylafax will then proceed as if it itself would have 
determined the type of call.

The issue I have is that when I do this I get a modem write error 
after calling the sys::write(fd,......   call and as I don't fully 
understand the inheritance rules that Sam applies and neither have I 
been able to understand the Dispatch class. I am at a loss in trying 
to guess what is going wrong.

This is why I am looking for any working example for simple and bare 
bones it is.

Thanks for your help,

Julio

Note: I hope this clarifies what egetty is.
> Date:          Mon, 20 Oct 1997 09:32:39 +0100
> To:            HylaFAX List <flexfax@sgi.com>
> From:          Phil Watkinson <pkw@elgro.co.uk>
> Subject:       egetty and vgetty
> Cc:            Julio C Spinelli <spine001@gold.tc.umn.edu>

> 
> On Fri, 17 Oct 1997, Julio C Spinelli wrote :
> >It would be great if you could add how to use egetty and give a very 
> >simple example that does nothing but recognized the call as a fax 
> >call. This will help a lot of people (including me). Same would be 
> >great for vgetty.
> 
> You have the advantage over me here, since I know nothing about egetty
> ( I'm only the piano player..). Like what does the 'e' suppose to stand
> for?  Q123 of the FAQ doesn't give much detail; Nico Garcia asked a 
> couple of weeks ago for the location of the egetty program, with no 
> reply; so I guess hard information about it maybe difficult to come by. 
> 
> A description of Vgetty and its homepage is mentioned on the contrib.html
> page, as part of the section on associated programs to HylaFAX. The
> TODO List (todo.html) also acknowledges that integration issues with
> vgetty are a high priority - it is required to implement a 'faxback'
> facility.
> 
> [ A faxback system is where a person calls in, hears a string of menus
>   which they navigate with the keypad of their phone. When they decide
>   on a document they want, the system faxes it back to the fax number
>   of their choice ].
> 
> If you are interested in doing some work in this area, you will find 
> more information by searching the mailing list archives for emails from
> 'Ivan Kohler' and 'Bret A. Schuhmacher'.
> 
> Cheers,
> Phil Watkinson,
> Boston, UK.
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Re: egetty and vgetty
To: spine001@gold.tc.umn.edu (Julio C Spinelli)
Date: Wed, 22 Oct 1997 17:28:00 +0200 (MSZ)
Cc: pkw@elgro.co.uk (Phil Watkinson), flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Julio C Spinelli wrote:
    
    Thanks for the info. egetty is nothing, it does not exist at all. The 
    only meaning of the string egetty is that if EGettyArgs is defined as 
    a non nul string and AnswerRotary is set to "extern". Hylafax will 
    call a program /bin/egetty. If that program does not exist...
    
    It is the user/system developer's responsibility to write that 
    program. This program's job is to answer the incoming call, determine 
    its type and then exit with a CALLTYPE_XXX (FAX, DATA, DONE, ERROR, 
    VOICE) code. Hylafax will then proceed as if it itself would have 
    determined the type of call.
    
    The issue I have is that when I do this I get a modem write error 
    after calling the sys::write(fd,......   call and as I don't fully 
    understand the inheritance rules that Sam applies and neither have I 
    been able to understand the Dispatch class. I am at a loss in trying 
    to guess what is going wrong.
    
    This is why I am looking for any working example for simple and bare 
    bones it is.
    
    Thanks for your help,
    
    Julio
    
    Note: I hope this clarifies what egetty is.

Try this small example. It works for me (don't blame me about
this C-code, it's not the way I'm writing real code :-))
It works for me and faxgetty(1M) receives the fax fine (see the log).

	matthias


/* change this define to the real device on your system */
#define LINE "/dev/ttyFN02" 
main()
{
	int fd;
	fd = open(LINE, 2);
	if(fd < 0)
		exit(0);
	write(fd, "ATA\r", 4);   /* time to pick up the call */
	close(fd);
	exit(2);
}

it gives the following session log:


Oct 22 19:10:12.67: [ 7560]: SESSION BEGIN 00003087 498961308189
Oct 22 19:10:12.74: [ 7560]: EXTERN GETTY: START "/bin/egetty test", pid 7798
Oct 22 19:10:12.98: [ 7560]: MODEM input buffering enabled
Oct 22 19:10:13.36: [ 7560]: EXTERN GETTY: exit status 01000
Oct 22 19:10:13.38: [ 7560]: ANSWER: FAX CONNECTION
Oct 22 19:10:13.39: [ 7560]: MODEM input buffering enabled
Oct 22 19:10:13.45: [ 7560]: RECV FAX: begin
Oct 22 19:10:30.07: [ 7560]: --> [11:CONNECT FAX]
Oct 22 19:10:30.08: [ 7560]: --> [4:+FCO]
Oct 22 19:10:32.59: [ 7560]: --> [26:+FTI:        498961308398 ]
Oct 22 19:10:32.59: [ 7560]: REMOTE TSI "498961308398"
Oct 22 19:10:32.60: [ 7560]: --> [20:+FCS:0,3,0,2,1,0,0,5]
Oct 22 19:10:32.61: [ 7560]: REMOTE wants 9600 bit/s
Oct 22 19:10:32.62: [ 7560]: REMOTE wants page width 1728 pixels in 215 mm
Oct 22 19:10:32.62: [ 7560]: REMOTE wants unlimited page length 
Oct 22 19:10:32.63: [ 7560]: REMOTE wants 3.85 line/mm
Oct 22 19:10:32.63: [ 7560]: REMOTE wants 2-D MR
Oct 22 19:10:32.64: [ 7560]: --> [2:OK]
Oct 22 19:10:32.73: [ 7560]: <-- [7:AT+FDR\r]
Oct 22 19:10:34.37: [ 7560]: --> [20:+FCS:0,3,0,2,1,0,0,5]
Oct 22 19:10:34.37: [ 7560]: REMOTE wants 9600 bit/s
Oct 22 19:10:34.38: [ 7560]: REMOTE wants page width 1728 pixels in 215 mm
Oct 22 19:10:34.38: [ 7560]: REMOTE wants unlimited page length 
Oct 22 19:10:34.39: [ 7560]: REMOTE wants 3.85 line/mm
Oct 22 19:10:34.40: [ 7560]: REMOTE wants 2-D MR
Oct 22 19:10:34.87: [ 7560]: --> [7:CONNECT]
Oct 22 19:10:34.87: [ 7560]: RECV: begin page
Oct 22 19:10:34.88: [ 7560]: RECV: send trigger 022
Oct 22 19:10:34.89: [ 7560]: <-- data [1]
Oct 22 19:11:04.60: [ 7560]: RECV: 13292 bytes of data, 1145 total lines
Oct 22 19:11:04.68: [ 7560]: --> [16:+FPS:1,474,0,0,0]
Oct 22 19:11:05.94: [ 7560]: --> [6:+FET:2]
Oct 22 19:11:05.95: [ 7560]: RECV recv EOP (no more pages or documents)
Oct 22 19:11:05.96: [ 7560]: --> [2:OK]
Oct 22 19:11:06.01: [ 7560]: RECV send MCF (message confirmation)
Oct 22 19:11:06.04: [ 7560]: RECV FAX (00003087): from 498961308398, page 1 in 0:34, INF, 3.85 line/mm, 2-D MR
Oct 22 19:11:06.13: [ 7560]: RECV FAX (00003087): recvq/fax00201.tif from 498961308398, route to <unspecified>, 1 pages in 0:53
Oct 22 19:11:06.20: [ 7560]: <-- [7:AT+FDR\r]
Oct 22 19:11:09.08: [ 7560]: --> [7:+FHS:00]
Oct 22 19:11:09.08: [ 7560]: REMOTE HANGUP: Normal and proper end of connection (code 0)
Oct 22 19:11:09.12: [ 7560]: RECV FAX: bin/faxrcvd "recvq/fax00201.tif" "ttyFN02" "00003087" ""
Oct 22 19:11:12.13: [ 7560]: RECV FAX: end
Oct 22 19:11:12.14: [ 7560]: SESSION END

From Matthias.Apitz@SOFTCON.de  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Re: egetty and vgetty
To: spine001@gold.tc.umn.edu (Julio C Spinelli)
Date: Wed, 22 Oct 1997 17:28:00 +0200 (MSZ)
Cc: pkw@elgro.co.uk (Phil Watkinson), flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps

Julio C Spinelli wrote:
    
    Thanks for the info. egetty is nothing, it does not exist at all. The 
    only meaning of the string egetty is that if EGettyArgs is defined as 
    a non nul string and AnswerRotary is set to "extern". Hylafax will 
    call a program /bin/egetty. If that program does not exist...
    
    It is the user/system developer's responsibility to write that 
    program. This program's job is to answer the incoming call, determine 
    its type and then exit with a CALLTYPE_XXX (FAX, DATA, DONE, ERROR, 
    VOICE) code. Hylafax will then proceed as if it itself would have 
    determined the type of call.
    
    The issue I have is that when I do this I get a modem write error 
    after calling the sys::write(fd,......   call and as I don't fully 
    understand the inheritance rules that Sam applies and neither have I 
    been able to understand the Dispatch class. I am at a loss in trying 
    to guess what is going wrong.
    
    This is why I am looking for any working example for simple and bare 
    bones it is.
    
    Thanks for your help,
    
    Julio
    
    Note: I hope this clarifies what egetty is.

Try this small example. It works for me (don't blame me about
this C-code, it's not the way I'm writing real code :-))
It works for me and faxgetty(1M) receives the fax fine (see the log).

	matthias


/* change this define to the real device on your system */
#define LINE "/dev/ttyFN02" 
main()
{
	int fd;
	fd = open(LINE, 2);
	if(fd < 0)
		exit(0);
	write(fd, "ATA\r", 4);   /* time to pick up the call */
	close(fd);
	exit(2);
}

it gives the following session log:


Oct 22 19:10:12.67: [ 7560]: SESSION BEGIN 00003087 498961308189
Oct 22 19:10:12.74: [ 7560]: EXTERN GETTY: START "/bin/egetty test", pid 7798
Oct 22 19:10:12.98: [ 7560]: MODEM input buffering enabled
Oct 22 19:10:13.36: [ 7560]: EXTERN GETTY: exit status 01000
Oct 22 19:10:13.38: [ 7560]: ANSWER: FAX CONNECTION
Oct 22 19:10:13.39: [ 7560]: MODEM input buffering enabled
Oct 22 19:10:13.45: [ 7560]: RECV FAX: begin
Oct 22 19:10:30.07: [ 7560]: --> [11:CONNECT FAX]
Oct 22 19:10:30.08: [ 7560]: --> [4:+FCO]
Oct 22 19:10:32.59: [ 7560]: --> [26:+FTI:        498961308398 ]
Oct 22 19:10:32.59: [ 7560]: REMOTE TSI "498961308398"
Oct 22 19:10:32.60: [ 7560]: --> [20:+FCS:0,3,0,2,1,0,0,5]
Oct 22 19:10:32.61: [ 7560]: REMOTE wants 9600 bit/s
Oct 22 19:10:32.62: [ 7560]: REMOTE wants page width 1728 pixels in 215 mm
Oct 22 19:10:32.62: [ 7560]: REMOTE wants unlimited page length 
Oct 22 19:10:32.63: [ 7560]: REMOTE wants 3.85 line/mm
Oct 22 19:10:32.63: [ 7560]: REMOTE wants 2-D MR
Oct 22 19:10:32.64: [ 7560]: --> [2:OK]
Oct 22 19:10:32.73: [ 7560]: <-- [7:AT+FDR\r]
Oct 22 19:10:34.37: [ 7560]: --> [20:+FCS:0,3,0,2,1,0,0,5]
Oct 22 19:10:34.37: [ 7560]: REMOTE wants 9600 bit/s
Oct 22 19:10:34.38: [ 7560]: REMOTE wants page width 1728 pixels in 215 mm
Oct 22 19:10:34.38: [ 7560]: REMOTE wants unlimited page length 
Oct 22 19:10:34.39: [ 7560]: REMOTE wants 3.85 line/mm
Oct 22 19:10:34.40: [ 7560]: REMOTE wants 2-D MR
Oct 22 19:10:34.87: [ 7560]: --> [7:CONNECT]
Oct 22 19:10:34.87: [ 7560]: RECV: begin page
Oct 22 19:10:34.88: [ 7560]: RECV: send trigger 022
Oct 22 19:10:34.89: [ 7560]: <-- data [1]
Oct 22 19:11:04.60: [ 7560]: RECV: 13292 bytes of data, 1145 total lines
Oct 22 19:11:04.68: [ 7560]: --> [16:+FPS:1,474,0,0,0]
Oct 22 19:11:05.94: [ 7560]: --> [6:+FET:2]
Oct 22 19:11:05.95: [ 7560]: RECV recv EOP (no more pages or documents)
Oct 22 19:11:05.96: [ 7560]: --> [2:OK]
Oct 22 19:11:06.01: [ 7560]: RECV send MCF (message confirmation)
Oct 22 19:11:06.04: [ 7560]: RECV FAX (00003087): from 498961308398, page 1 in 0:34, INF, 3.85 line/mm, 2-D MR
Oct 22 19:11:06.13: [ 7560]: RECV FAX (00003087): recvq/fax00201.tif from 498961308398, route to <unspecified>, 1 pages in 0:53
Oct 22 19:11:06.20: [ 7560]: <-- [7:AT+FDR\r]
Oct 22 19:11:09.08: [ 7560]: --> [7:+FHS:00]
Oct 22 19:11:09.08: [ 7560]: REMOTE HANGUP: Normal and proper end of connection (code 0)
Oct 22 19:11:09.12: [ 7560]: RECV FAX: bin/faxrcvd "recvq/fax00201.tif" "ttyFN02" "00003087" ""
Oct 22 19:11:12.13: [ 7560]: RECV FAX: end
Oct 22 19:11:12.14: [ 7560]: SESSION END

From spine001@gold.tc.umn.edu  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 17:07:54 -0500 (CDT)
From: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
To: Matthias Apitz <Matthias.Apitz@SOFTCON.de>
cc: Phil Watkinson <pkw@elgro.co.uk>, flexfax@sgi.com
Subject: Re: flexfax: Re: egetty and vgetty

Thanks very much for the example, the only difference between what you are
doing and what I was doing is that you are issuing ATA before exiting
egetty. 100 to 1 that was my mistake, now I realize that of course hylafax
should expect that start the fax call handling is the responsibility of
the egetty program. And of course egetty should also reinitialize the
modem before issuing the ATA command. 

If this works I will still read the changes sent to me and try to see what
he did. I'll also send you a few lines with the changes that I recommend,

Thanks very much!

Julio

On Wed, 22 Oct 1997 Matthias.Apitz@SOFTCON.de wrote:

> Julio C Spinelli wrote:
>     
>     Thanks for the info. egetty is nothing, it does not exist at all. The 
>     only meaning of the string egetty is that if EGettyArgs is defined as 
>     a non nul string and AnswerRotary is set to "extern". Hylafax will 
>     call a program /bin/egetty. If that program does not exist...
>     
>     It is the user/system developer's responsibility to write that 
>     program. This program's job is to answer the incoming call, determine 
>     its type and then exit with a CALLTYPE_XXX (FAX, DATA, DONE, ERROR, 
>     VOICE) code. Hylafax will then proceed as if it itself would have 
>     determined the type of call.
>     
>     The issue I have is that when I do this I get a modem write error 
>     after calling the sys::write(fd,......   call and as I don't fully 
>     understand the inheritance rules that Sam applies and neither have I 
>     been able to understand the Dispatch class. I am at a loss in trying 
>     to guess what is going wrong.
>     
>     This is why I am looking for any working example for simple and bare 
>     bones it is.
>     
>     Thanks for your help,
>     
>     Julio
>     
>     Note: I hope this clarifies what egetty is.
> 
> Try this small example. It works for me (don't blame me about
> this C-code, it's not the way I'm writing real code :-))
> It works for me and faxgetty(1M) receives the fax fine (see the log).
> 
> 	matthias
> 
> 
> /* change this define to the real device on your system */
> #define LINE "/dev/ttyFN02" 
> main()
> {
> 	int fd;
> 	fd = open(LINE, 2);
> 	if(fd < 0)
> 		exit(0);
> 	write(fd, "ATA\r", 4);   /* time to pick up the call */
> 	close(fd);
> 	exit(2);
> }
> 
> it gives the following session log:
> 
> 
> Oct 22 19:10:12.67: [ 7560]: SESSION BEGIN 00003087 498961308189
> Oct 22 19:10:12.74: [ 7560]: EXTERN GETTY: START "/bin/egetty test", pid 7798
> Oct 22 19:10:12.98: [ 7560]: MODEM input buffering enabled
> Oct 22 19:10:13.36: [ 7560]: EXTERN GETTY: exit status 01000
> Oct 22 19:10:13.38: [ 7560]: ANSWER: FAX CONNECTION
> Oct 22 19:10:13.39: [ 7560]: MODEM input buffering enabled
> Oct 22 19:10:13.45: [ 7560]: RECV FAX: begin
> Oct 22 19:10:30.07: [ 7560]: --> [11:CONNECT FAX]
> Oct 22 19:10:30.08: [ 7560]: --> [4:+FCO]
> Oct 22 19:10:32.59: [ 7560]: --> [26:+FTI:        498961308398 ]
> Oct 22 19:10:32.59: [ 7560]: REMOTE TSI "498961308398"
> Oct 22 19:10:32.60: [ 7560]: --> [20:+FCS:0,3,0,2,1,0,0,5]
> Oct 22 19:10:32.61: [ 7560]: REMOTE wants 9600 bit/s
> Oct 22 19:10:32.62: [ 7560]: REMOTE wants page width 1728 pixels in 215 mm
> Oct 22 19:10:32.62: [ 7560]: REMOTE wants unlimited page length 
> Oct 22 19:10:32.63: [ 7560]: REMOTE wants 3.85 line/mm
> Oct 22 19:10:32.63: [ 7560]: REMOTE wants 2-D MR
> Oct 22 19:10:32.64: [ 7560]: --> [2:OK]
> Oct 22 19:10:32.73: [ 7560]: <-- [7:AT+FDR\r]
> Oct 22 19:10:34.37: [ 7560]: --> [20:+FCS:0,3,0,2,1,0,0,5]
> Oct 22 19:10:34.37: [ 7560]: REMOTE wants 9600 bit/s
> Oct 22 19:10:34.38: [ 7560]: REMOTE wants page width 1728 pixels in 215 mm
> Oct 22 19:10:34.38: [ 7560]: REMOTE wants unlimited page length 
> Oct 22 19:10:34.39: [ 7560]: REMOTE wants 3.85 line/mm
> Oct 22 19:10:34.40: [ 7560]: REMOTE wants 2-D MR
> Oct 22 19:10:34.87: [ 7560]: --> [7:CONNECT]
> Oct 22 19:10:34.87: [ 7560]: RECV: begin page
> Oct 22 19:10:34.88: [ 7560]: RECV: send trigger 022
> Oct 22 19:10:34.89: [ 7560]: <-- data [1]
> Oct 22 19:11:04.60: [ 7560]: RECV: 13292 bytes of data, 1145 total lines
> Oct 22 19:11:04.68: [ 7560]: --> [16:+FPS:1,474,0,0,0]
> Oct 22 19:11:05.94: [ 7560]: --> [6:+FET:2]
> Oct 22 19:11:05.95: [ 7560]: RECV recv EOP (no more pages or documents)
> Oct 22 19:11:05.96: [ 7560]: --> [2:OK]
> Oct 22 19:11:06.01: [ 7560]: RECV send MCF (message confirmation)
> Oct 22 19:11:06.04: [ 7560]: RECV FAX (00003087): from 498961308398, page 1 in 0:34, INF, 3.85 line/mm, 2-D MR
> Oct 22 19:11:06.13: [ 7560]: RECV FAX (00003087): recvq/fax00201.tif from 498961308398, route to <unspecified>, 1 pages in 0:53
> Oct 22 19:11:06.20: [ 7560]: <-- [7:AT+FDR\r]
> Oct 22 19:11:09.08: [ 7560]: --> [7:+FHS:00]
> Oct 22 19:11:09.08: [ 7560]: REMOTE HANGUP: Normal and proper end of connection (code 0)
> Oct 22 19:11:09.12: [ 7560]: RECV FAX: bin/faxrcvd "recvq/fax00201.tif" "ttyFN02" "00003087" ""
> Oct 22 19:11:12.13: [ 7560]: RECV FAX: end
> Oct 22 19:11:12.14: [ 7560]: SESSION END
> 

From Matthias.Apitz@SOFTCON.de  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Re: egetty and vgetty
To: spine001@gold.tc.umn.edu (Prof. Julio C. Spinelli)
Date: Thu, 23 Oct 1997 07:51:56 +0200 (MSZ)
Cc: flexfax@sgi.com, pkw@elgro.co.uk (Phil Watkinson)
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps

Prof. Julio C. Spinelli wrote:
    
    Thanks very much for the example, the only difference between what you are
    doing and what I was doing is that you are issuing ATA before exiting
    egetty. 100 to 1 that was my mistake, now I realize that of course hylafax
    should expect that start the fax call handling is the responsibility of
    the egetty program. And of course egetty should also reinitialize the
    modem before issuing the ATA command. 

I don't think that the called egetty has to reinitialize
the modem. The logic goes IMHO that way:

- faxgetty(1M) sees the modem ringing (RING) and is configured
  to use egetty to let egetty deduce what's to do with this call
- faxgetty(1M) handover the call (keep in mind the call and the
  caller is waiting)
- the first think that egetty hast to do: investigate the RING
  message from the modem (there are modems showing the call-type
  in the RING message) or pick-up this call with ATA (or something
  like that) ond decide what's to do with this answered call,
  e.g. look for PPP packets after the two modem are CONNECT'ed
- ...
- tell later faxgetty(1M) with the exit-state what's to do now
  with this (maybe done) call

reinitializing the modem will perhaps drop that call;

	matthias
    
    If this works I will still read the changes sent to me and try to see what
    he did. I'll also send you a few lines with the changes that I recommend,
    
    Thanks very much!
    
    Julio

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Re: egetty and vgetty
To: spine001@gold.tc.umn.edu (Prof. Julio C. Spinelli)
Date: Thu, 23 Oct 1997 07:51:56 +0200 (MSZ)
Cc: flexfax@sgi.com, pkw@elgro.co.uk (Phil Watkinson)
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Prof. Julio C. Spinelli wrote:
    
    Thanks very much for the example, the only difference between what you are
    doing and what I was doing is that you are issuing ATA before exiting
    egetty. 100 to 1 that was my mistake, now I realize that of course hylafax
    should expect that start the fax call handling is the responsibility of
    the egetty program. And of course egetty should also reinitialize the
    modem before issuing the ATA command. 

I don't think that the called egetty has to reinitialize
the modem. The logic goes IMHO that way:

- faxgetty(1M) sees the modem ringing (RING) and is configured
  to use egetty to let egetty deduce what's to do with this call
- faxgetty(1M) handover the call (keep in mind the call and the
  caller is waiting)
- the first think that egetty hast to do: investigate the RING
  message from the modem (there are modems showing the call-type
  in the RING message) or pick-up this call with ATA (or something
  like that) ond decide what's to do with this answered call,
  e.g. look for PPP packets after the two modem are CONNECT'ed
- ...
- tell later faxgetty(1M) with the exit-state what's to do now
  with this (maybe done) call

reinitializing the modem will perhaps drop that call;

	matthias
    
    If this works I will still read the changes sent to me and try to see what
    he did. I'll also send you a few lines with the changes that I recommend,
    
    Thanks very much!
    
    Julio

From spine001@gold.tc.umn.edu  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 13:59:43 -0500 (CDT)
From: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
To: Matthias Apitz <Matthias.Apitz@SOFTCON.de>
cc: flexfax@sgi.com, Phil Watkinson <pkw@elgro.co.uk>
Subject: Re: flexfax: Re: egetty and vgetty

Dear Matthias,

When I followed the code I found that:

1) answerCall is called
2) if Answer Rotary is set to "extern"
3) if EGettyArgs tag is defined non-null in the config.xxx file
4) answerCall calls runGetty that in turn return with the type of call
5) when runGetty returns if if returns with CALLTYPE_DONE nothing is done
6) if the call returns with CALLTYPE != ERROR then the
modemAnswerCallCmd is called. /* the guy who changed the code here calls
modemAnswerCall*/
7) modemAnswerCallCmd just calls answerCallCmd.
8) answerCallCmd defined in ClassModem for CALLTYPE_FAX sets beginCmd to
conf.answerFaxBeginCmd and then sends that command to the modem.
9) answerFaxBeginCmd is defined in the atcmds[] structure in the
ModemConfig class to be the whatever is assigned to the tag
ModemAnswerFaxBeginCmd in the config.xxx file.

This tag is used to define baud rates for modems that switch to fixed baud
rates when they are in fax mode like the Rockwell based ones, a typical
one would be "<19200><xon>". By this time hylafax thinks that the carrier
has been stablished and it is ready to receive the fax.

Funny enough, my egetty does not even when I incorporated the ata\r before
exiting. OK my egetty works but when it returns to faxgetty it hangs up.
On a private email I'll send to you the important part of the strace
output of the faxgetty command.

The guy that changed hylafax did two things, first intead of calling
modemAnswerCallCmd, calls modemAnswerCall, which does answer the call,
meaning you can leave egetty without having answered. [But I don't want to
do that for me it is more efficient to answer the call in egetty]

The second thing that he did is to disable the execution of the following 
statments:

signal(SIGTERM, fxSIGHANDLER(SIG_DFL));
signal(SIGHUP, fxSIGHANDLER(sigHUP));
setupSession(fd);

in Getty::run(int fd, fxBool parentIsInit) function when argv[0]==egetty
(I know this is the wrong syntaxis...)

I don't have a clue of the implications of doing this. I don't even
understand what these signals are for, except may be for SIGTERM that I
think catches Ctrl C.

With respect to setupSession, it seems to change the ownership of the
modem file descriptor fd and then I get really confused, it seems that it
tries to duplicate the modem file descriptor into the standard output file
descriptor. Would this make any output intended for the screen go to the
modem?? And then it tries to duplicate the modem file descriptor to be the
STDERR_FILENO, standard error. Could you anybody clarify this? The lines
of code I am talking about are:

(dup2(fd, STDOUT_FILENO); and (dup2(fd,STDERR_FILENO). Eventhough I don't
understand exactly what they do, they may make sense for the getty
process, but why do this (what I guess it does) for egetty and vgetty?

Anyway, all this is to say that:

1) If you use the egetty hook as it is you need to answer the call at the
end of your egetty code.
2) If you change the modem settings, don't expect faxGetty to reset your
modem to fax mode. You need to do it yourself inside egetty.
3) It does not work yet for me, hence I am missing something, may be it is
in the changes that my pal made. Can somebody help me understand what the
implications of those changes are (not the modemAnserCallCmd, that one I
got)?
4) This may be a Linux only problem.

Thanks for your help in advance,

Julio

On Thu, 23 Oct 1997 Matthias.Apitz@SOFTCON.de wrote:

> Prof. Julio C. Spinelli wrote:
>     
>     Thanks very much for the example, the only difference between what you are
>     doing and what I was doing is that you are issuing ATA before exiting
>     egetty. 100 to 1 that was my mistake, now I realize that of course hylafax
>     should expect that start the fax call handling is the responsibility of
>     the egetty program. And of course egetty should also reinitialize the
>     modem before issuing the ATA command. 
> 
> I don't think that the called egetty has to reinitialize
> the modem. The logic goes IMHO that way:
> 
> - faxgetty(1M) sees the modem ringing (RING) and is configured
>   to use egetty to let egetty deduce what's to do with this call
> - faxgetty(1M) handover the call (keep in mind the call and the
>   caller is waiting)
> - the first think that egetty hast to do: investigate the RING
>   message from the modem (there are modems showing the call-type
>   in the RING message) or pick-up this call with ATA (or something
>   like that) ond decide what's to do with this answered call,
>   e.g. look for PPP packets after the two modem are CONNECT'ed
> - ...
> - tell later faxgetty(1M) with the exit-state what's to do now
>   with this (maybe done) call
> 
> reinitializing the modem will perhaps drop that call;
> 
> 	matthias
>     
>     If this works I will still read the changes sent to me and try to see what
>     he did. I'll also send you a few lines with the changes that I recommend,
>     
>     Thanks very much!
>     
>     Julio
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 29 Oct 1997 14:46:05 -0800 (PST)
From: Camden Clarke <cclarke@inconceivable.com>
To: flexfax@sgi.com
Subject: Re: flexfax: Re: egetty and vgetty
Sender: owner-flexfax@celestial.com


You don't actually have to open the modem in your egetty program; HylaFAX
will leave you with stdin and stdout setup to the modem.  This is intentional,
so that egetty doesn't have to know where to find the modem, or deal
with opening it correctly for shared access (e.g. Suns).
                -Cam Clarke
 
In astatos.flex-fax you write:
 
>Julio C Spinelli wrote:
>
>    Thanks for the info. egetty is nothing, it does not exist at all. The
>    only meaning of the string egetty is that if EGettyArgs is defined as
>    a non nul string and AnswerRotary is set to "extern". Hylafax will
>    call a program /bin/egetty. If that program does not exist...
>
>    It is the user/system developer's responsibility to write that
>    program. This program's job is to answer the incoming call, determine
>    its type and then exit with a CALLTYPE_XXX (FAX, DATA, DONE, ERROR,
>    VOICE) code. Hylafax will then proceed as if it itself would have
>    determined the type of call.
>
>    The issue I have is that when I do this I get a modem write error
>    after calling the sys::write(fd,......   call and as I don't fully
>    understand the inheritance rules that Sam applies and neither have I
>    been able to understand the Dispatch class. I am at a loss in trying
>    to guess what is going wrong.
>
>    This is why I am looking for any working example for simple and bare
>    bones it is.
>
>    Thanks for your help,
>
>    Julio
>
>    Note: I hope this clarifies what egetty is.
 
>Try this small example. It works for me (don't blame me about
>this C-code, it's not the way I'm writing real code :-))
>It works for me and faxgetty(1M) receives the fax fine (see the log).
 
>   matthias
 
 
>/* change this define to the real device on your system */
>#define LINE "/dev/ttyFN02"
>main()
>{
>   int fd;
>   fd = open(LINE, 2);
>   if(fd < 0)
>       exit(0);
>   write(fd, "ATA\r", 4);   /* time to pick up the call */
>   close(fd);
>   exit(2);
>}

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Re: egetty and vgetty
To: cclarke@inconceivable.com (Camden Clarke)
Date: Thu, 30 Oct 1997 09:58:18 +0100 (MEZ)
Cc: flexfax@sgi.com, spine001@gold.tc.umn.edu (Julio C Spinelli)
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Camden Clarke wrote:
    
    
    You don't actually have to open the modem in your egetty program; HylaFAX
    will leave you with stdin and stdout setup to the modem.  This is intentional,
    so that egetty doesn't have to know where to find the modem, or deal
    with opening it correctly for shared access (e.g. Suns).
                    -Cam Clarke

You're pretty much correct and my code below was only a small
and dirty example (which works on my system too). I added this
re-open because Julio has a patch for HylaFAX which closes the
modem-Fd before handing the modem over to the child. Don't know
why someone did this patch and at the moment I think this
is not the way the things should changed, Julio?

	matthias
     
    In astatos.flex-fax you write:
     
    >Julio C Spinelli wrote:
    >
    >    Thanks for the info. egetty is nothing, it does not exist at all. The
    >    only meaning of the string egetty is that if EGettyArgs is defined as
    >    a non nul string and AnswerRotary is set to "extern". Hylafax will
    >    call a program /bin/egetty. If that program does not exist...
    >
    >    It is the user/system developer's responsibility to write that
    >    program. This program's job is to answer the incoming call, determine
    >    its type and then exit with a CALLTYPE_XXX (FAX, DATA, DONE, ERROR,
    >    VOICE) code. Hylafax will then proceed as if it itself would have
    >    determined the type of call.
    >
    >    The issue I have is that when I do this I get a modem write error
    >    after calling the sys::write(fd,......   call and as I don't fully
    >    understand the inheritance rules that Sam applies and neither have I
    >    been able to understand the Dispatch class. I am at a loss in trying
    >    to guess what is going wrong.
    >
    >    This is why I am looking for any working example for simple and bare
    >    bones it is.
    >
    >    Thanks for your help,
    >
    >    Julio
    >
    >    Note: I hope this clarifies what egetty is.
     
    >Try this small example. It works for me (don't blame me about
    >this C-code, it's not the way I'm writing real code :-))
    >It works for me and faxgetty(1M) receives the fax fine (see the log).
     
    >   matthias
     
     
    >/* change this define to the real device on your system */
    >#define LINE "/dev/ttyFN02"
    >main()
    >{
    >   int fd;
    >   fd = open(LINE, 2);
    >   if(fd < 0)
    >       exit(0);
    >   write(fd, "ATA\r", 4);   /* time to pick up the call */
    >   close(fd);
    >   exit(2);
    >}

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Comments: Authenticated sender is <spine001@gold.tc.umn.edu>
From: Julio C Spinelli <spine001@gold.tc.umn.edu>
To: Matthias.Apitz@SOFTCON.de
Date: Thu, 30 Oct 1997 14:36:59 +0000
Subject: Re: flexfax: Re: egetty and vgetty
CC: flexfax@sgi.com,
        Julio C Spinelli <spine001@gold.tc.umn.edu> (Julio C Spinelli)
Sender: owner-flexfax@celestial.com

Dear Matthias and Camdem,

I am afraid that trying to clarify, and due to my lack of 
understanding I introduced some confusion. First, thanks Camdem for 
the clarification, now I know what the dup2 statements are doing. 

1) The business of closing the modem fd is what I was told by the 
author of the patch. When I checked the code that was not what he was 
doing (probably his memory failed). (See my previous email on the 
subject)

2) The problem seems to be that in Linux, for whatever the reason, 
when egetty returns control to the parent process, the output from 
the modem gets lost (as you saw in the strace file I sent you). There 
is also a write(fd, ...) errno 5 reported. And Hylafax hangs up the 
call without answering in fax mode (when returning calltype_fax).

3) This problem can be easily corrected by having hylafax issue the 
ata command after egetty returns.

A question, would anybody explain to me why in Gettt.c++ it is not 
necessary to dup2 the stdin file descriptor and only the stdout and 
etderr??? to redirect the standard input and output to the fd of the 
modem.

Thanks in advance,

Julio

> From:          Matthias.Apitz@SOFTCON.de
> Subject:       Re: flexfax: Re: egetty and vgetty
> To:            cclarke@inconceivable.com (Camden Clarke)
> Date:          Thu, 30 Oct 1997 09:58:18 +0100 (MEZ)
> Cc:            flexfax@sgi.com,
>                Julio C Spinelli <spine001@gold.tc.umn.edu> (Julio C Spinelli)
> Reply-to:      Matthias.Apitz@SOFTCON.de (Matthias Apitz)

> Camden Clarke wrote:
>     
>     
>     You don't actually have to open the modem in your egetty program; HylaFAX
>     will leave you with stdin and stdout setup to the modem.  This is intentional,
>     so that egetty doesn't have to know where to find the modem, or deal
>     with opening it correctly for shared access (e.g. Suns).
>                     -Cam Clarke
> 
> You're pretty much correct and my code below was only a small
> and dirty example (which works on my system too). I added this
> re-open because Julio has a patch for HylaFAX which closes the
> modem-Fd before handing the modem over to the child. Don't know
> why someone did this patch and at the moment I think this
> is not the way the things should changed, Julio?
> 
> 	matthias
>      
>     In astatos.flex-fax you write:
>      
>     >Julio C Spinelli wrote:
>     >
>     >    Thanks for the info. egetty is nothing, it does not exist at all. The
>     >    only meaning of the string egetty is that if EGettyArgs is defined as
>     >    a non nul string and AnswerRotary is set to "extern". Hylafax will
>     >    call a program /bin/egetty. If that program does not exist...
>     >
>     >    It is the user/system developer's responsibility to write that
>     >    program. This program's job is to answer the incoming call, determine
>     >    its type and then exit with a CALLTYPE_XXX (FAX, DATA, DONE, ERROR,
>     >    VOICE) code. Hylafax will then proceed as if it itself would have
>     >    determined the type of call.
>     >
>     >    The issue I have is that when I do this I get a modem write error
>     >    after calling the sys::write(fd,......   call and as I don't fully
>     >    understand the inheritance rules that Sam applies and neither have I
>     >    been able to understand the Dispatch class. I am at a loss in trying
>     >    to guess what is going wrong.
>     >
>     >    This is why I am looking for any working example for simple and bare
>     >    bones it is.
>     >
>     >    Thanks for your help,
>     >
>     >    Julio
>     >
>     >    Note: I hope this clarifies what egetty is.
>      
>     >Try this small example. It works for me (don't blame me about
>     >this C-code, it's not the way I'm writing real code :-))
>     >It works for me and faxgetty(1M) receives the fax fine (see the log).
>      
>     >   matthias
>      
>      
>     >/* change this define to the real device on your system */
>     >#define LINE "/dev/ttyFN02"
>     >main()
>     >{
>     >   int fd;
>     >   fd = open(LINE, 2);
>     >   if(fd < 0)
>     >       exit(0);
>     >   write(fd, "ATA\r", 4);   /* time to pick up the call */
>     >   close(fd);
>     >   exit(2);
>     >}
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 20 Oct 1997 12:10:20 +0200
From: Jean-Rene Beauvais <beauvais@irisa.fr>
Organization: IRISA, Campus de Beaulieu, 35042 Rennes Cedex, FRANCE
To: flexfax@sgi.com
Subject: flexfax: /dev/socksys
X-MIME-Autoconverted: from 8bit to quoted-printable by cheyenne.irisa.fr id MAA00999
Sender: owner-flexfax@celestial.com

Bonjour,

I installed Hylafax v4.0 on sco unix ver4.2 and i have the following
problem when running the daemons (using faxsetup) or use faxstat:

libsocket: open(/dev/socksys) failure:
No such file or directory

What is this device that I don't have ? Where could I get it ?

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Jean-Ren Beauvais               |  http://www.irisa.fr/prive/beauvais/
IRISA / INRIA, projet EP-ATR     |             phone: +33 2 99 84 75 99
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
X-Authentication-Warning: tui.pinnacle.co.nz: jonc owned process doing -bs
Date: Tue, 21 Oct 1997 09:21:53 +1300 (NZDT)
From: Jonathan Chen <jonc@pinnacle.co.nz>
To: Jean-Rene Beauvais <beauvais@irisa.fr>
cc: flexfax@sgi.com
Subject: Re: flexfax: /dev/socksys
Sender: owner-flexfax@celestial.com

On Mon, 20 Oct 1997, Jean-Rene Beauvais wrote:

> Bonjour,
> 
> I installed Hylafax v4.0 on sco unix ver4.2 and i have the following
> problem when running the daemons (using faxsetup) or use faxstat:
> 
> libsocket: open(/dev/socksys) failure:
> No such file or directory

For some reason your installation hasn't created a required device
file /var/spool/fax/dev/socksys. You can manually create the file
using mknod(S). Check out the device's type, major/minor numbers from
/dev/socksys and create one within /var/spool/fax/dev.
--
Jonathan Chen                          e-mail : jonc@pinnacle.co.nz
  Pinnacle Software Ltd                Voice  : +64.9.415.4460
  Auckland, New Zealand                Fax    : +64.9.415.4250

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
X-Authentication-Warning: tui.pinnacle.co.nz: jonc owned process doing -bs
Date: Wed, 22 Oct 1997 08:49:18 +1300 (NZDT)
From: Jonathan Chen <jonc@pinnacle.co.nz>
To: Jean-Rene Beauvais <beauvais@irisa.fr>
cc: Hylafax Mailing list <flexfax@sgi.com>
Subject: Re: flexfax: /dev/socksys
Sender: owner-flexfax@celestial.com

On Tue, 21 Oct 1997, Jean-Rene Beauvais wrote:

> Jonathan Chen wrote:
> 
> > For some reason your installation hasn't created a required device
> > file /var/spool/fax/dev/socksys. You can manually create the file
> > using mknod(S). Check out the device's type, major/minor numbers from
> > /dev/socksys and create one within /var/spool/fax/dev.
> 
> Thank you for your answer,
> 
> I don't have any /dev/socksys !! Do you know if I should have one ? If
> you are running SCO ver4.2, could you check for me and give me the
> major/minor numbers

Unfortunately, we're running 3.2v5.0.x here. I take it that the
SCO Networking Package has been installed on your system?
--
Jonathan Chen <jonc@pinnacle.co.nz>                 Once is dumb luck.
                                                 Twice is coincidence.
             Three times and Somebody Is Trying To Tell You Something.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Comments: Authenticated sender is <nmc@pop.unb.ca>
From: "Nicola Carter" <nmc@unb.ca>
Organization: UNB
To: flexfax@celestial.com
Date: Mon, 20 Oct 1997 16:06:36 -0300
Subject: flexfax: Solaris 2.5 Modem Replacement for USR
Reply-to: nmc@unb.ca
Sender: owner-flexfax@celestial.com

We are running Hylafax 4.0 PL 1.0 under Solaris 2.5. We have 
experienced the USR Sportster Class 2 Transmit Phase B error, and are 
currently running with an old USR Sportster 14,400 in Class 1 mode 
with 50-60% success (failure to retrain after the cover page causes 
3-4 blank pages to spew out at the destination before the rest of the 
fax oes through). A SUPRA FAX V.32 bis modem did not recover from 
that retrain after the first page.  We have experimented with flow 
control to no avail.

We could hobble along with the USR 14,400, putting up with the extra 
pages during retrain, for a while. Do any of you have a modem 
recommendation for this configuration? Or conversely, a 
modem-not-to-use? I notice that the pages with happy-face modem 
recommendations are somewhat out-of-date.

Nicola

<----nmc@unb.ca------------------- http://www.unb.ca -------->
Nicola Carter, Computing Services,University of New Brunswick
PO Box 4400, Fredericton, N.B. E3B 5A3 CANADA
Tel: 1-506-447-3031  Fax: 1-506-453-3590
  

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 20 Oct 1997 16:28:00 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Nicola Carter <nmc@unb.ca>
Cc: flexfax@celestial.com
Subject: Re: flexfax: Solaris 2.5 Modem Replacement for USR
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Mon, 20 Oct 1997, Nicola Carter wrote:

> We are running Hylafax 4.0 PL 1.0 under Solaris 2.5. We have 
> experienced the USR Sportster Class 2 Transmit Phase B error, and are 
> currently running with an old USR Sportster 14,400 in Class 1 mode 
> with 50-60% success (failure to retrain after the cover page causes 
> 3-4 blank pages to spew out at the destination before the rest of the 
> fax oes through). A SUPRA FAX V.32 bis modem did not recover from 
> that retrain after the first page.  We have experimented with flow 
> control to no avail.
> 
> We could hobble along with the USR 14,400, putting up with the extra 
> pages during retrain, for a while. Do any of you have a modem 
> recommendation for this configuration? Or conversely, a 
> modem-not-to-use? I notice that the pages with happy-face modem 
> recommendations are somewhat out-of-date.

You gain little from using 33.6 modems for fax. The higher baud rates are
for *data*, not fax....


			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNEu+yT/+ItycgIJRAQHFMwP/T/4u9jUR66IuYJX+KDj72E8hA1C4029g
11NUbzAXIwu4l9oCTnciZC8EK/UfMXQHglj0Idbcf6y13q7gFLyhZ4/O5Hnsfmfb
NaQYlh3YUkKV70p7GcC3CW+Q0LIk1btikH+XwZtZvuInjxeZCPrtRqX4jumiSept
9Y9MrXp5kfk=
=iySa
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
To: flexfax@sgi.com
Subject: flexfax: doc bug
Date: Sun, 19 Oct 1997 20:14:36 -0400
From: Trevor Blackwell <tlb@viaweb.com>
Sender: owner-flexfax@celestial.com


In 4.0pl1, man 5 config says:

 Tag            Type       Default          Description
[...]
 MaxDials-      integer    unlimited        max phone calls to make to transmit a job

but the default value is actually 12, taken from FAX_REDIALS
(config.h) through hfaxd/Jobs.c++.

Currently, it's puzzling that sendfax -T 50 still gives up after 12
tries. You have to specify a value for maxtries in ~fax/etc/config to
make this work, but this this isn't suggested by the man page. The
code probably does the right thing - the documentation should be
changed.

Don't ask why I need maxdials=50.

--
Trevor Blackwell         tlb@viaweb.com          (617) 876-2692 x982
http://www.viaweb.com/

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 16 Oct 1997 19:31:30 -0700
From: Chris Che <cw.cans@earthlink.net>
To: flexfax@sgi.com
Subject: flexfax: Concurrent modem dialing ?
Sender: owner-flexfax@celestial.com

Hello all,

I am doubting that my HylaFax setup is indeed utilizing all the modems (ttys)
that I have configured for it.

When I sent 2 faxes back to back, they all seem to queue up on the first TTY
port. When I use 'faxstat -s ', the second fax being block by the first one on
the same TTY.


Is it normal ? How can I fix it ?
thanks

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 17 Oct 1997 10:49:19 -0500 (EST)
From: Dave Lewis <dlewis@appn.ci.in.ameritech.com>
To: flexfax@sgi.com
Subject: flexfax: modem priority
Sender: owner-flexfax@celestial.com

Hi all.
I'm getting confusing results when setting modem priorities.  Our server 
has 9 modems, and in the config files /var/spool/fax/etc/config.tty#,
I've placed an individual 'ModemPriority:  ###' statement, as follows:
for tty5, /var/spool/fax/etc/config.tty5, which I want to have the highest 
priority: 
ModemPriority:	200
for tty6,
ModemPriority:	180  
etc. up to the last port, tty15, /var/spool/fax/etc/config.tty15
ModemPriority:	20

This is intended for outbound faxes only (almost all our faxes are such)
I restarted faxq, and I even restarted each faxgetty process (though as I 
stated, we don't receive many faxes on these lines), and the scheduler 
'kind-of' follows the priority I've set, but not exactly.  tty5,6 do seem 
to have the highest priority, but then followed by tty11, and tty15 
(which should have the lowest).  I'm a bit stumped here.  Is there some 
other magic going on?  Any insight about similar battles fought is 
appreciated.  Thanks, Dave Lewis

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Pierre MEURICE <P.Meurice@St-Emilion.citb.bull.net>
To: "'liste de diffusion hylafax'" <flexfax@sgi.com>
Subject: flexfax: flexflax: help busy problem with modem hayes accura
Date: Tue, 21 Oct 1997 09:48:19 +-100
Sender: owner-flexfax@celestial.com


Hello everibody,

I run hylafax on aix 4.X with a modem Hayes accura 28.8 model 5337,
and when we send a fax it always answer line is busy.

Please can you help us.
Thanck you.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 21 Oct 1997 13:26:47 -0500 (CDT)
From: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
To: Pierre MEURICE <P.Meurice@St-Emilion.citb.bull.net>
cc: "'liste de diffusion hylafax'" <flexfax@sgi.com>
Subject: Re: flexfax: flexflax: help busy problem with modem hayes accura
Sender: owner-flexfax@celestial.com

First test is to replace your modem with another one of the same brand.
Though, is has been mentioned that the Hayes Acura may not hang up under
all circumstance, which could be a problem,

Julio

On Tue, 21 Oct 1997, Pierre MEURICE wrote:

> 
> Hello everibody,
> 
> I run hylafax on aix 4.X with a modem Hayes accura 28.8 model 5337,
> and when we send a fax it always answer line is busy.
> 
> Please can you help us.
> Thanck you.
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 13:49:54 +0800 (HKT)
From: Super-User <root@peach.asiainter.net>
To: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
cc: Pierre MEURICE <P.Meurice@St-Emilion.citb.bull.net>,
        "'liste de diffusion hylafax'" <flexfax@sgi.com>
Subject: Re: flexfax: flexflax: help busy problem with modem hayes accura
Sender: owner-flexfax@celestial.com


On Tue, 21 Oct 1997, Prof. Julio C. Spinelli wrote:
> First test is to replace your modem with another one of the same brand.
> Though, is has been mentioned that the Hayes Acura may not hang up under
> all circumstance, which could be a problem,

i've found hayes (accura) modems to be very poor quality.. in data mode,
zoltrix brand internal modems especially have only a 10-20% chance of
connecting successfully.  they also lose their nvram settings when the
power is removed and the modem is turned on.

stay away from hayes modems if you can.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Pierre MEURICE <P.Meurice@St-Emilion.citb.bull.net>
To: "'liste de diffusion hylafax'" <flexfax@sgi.com>
Subject: flexfax: help !!! problem with modem hayes accura
Date: Wed, 22 Oct 1997 09:40:16 +-100
Sender: owner-flexfax@celestial.com

I' have test my Hylafax server with another modem hayes and it doesn't take the line, I have also tested it with a modem from another brand and it works.

It's a greet problem !!!

Can someone that use a modem Hayes accura 288 (V.34/V.FC+FAX) old or new model send me his modem configuring file : config.tty???? or the file  used to build it from in the config directory.

Thanck you to all



----------
De : 	Prof. Julio C. Spinelli[SMTP:spine001@gold.tc.umn.edu]
Date d'envoi:	mardi 21 octobre 1997 14:26
A:	Pierre MEURICE
Cc:	'liste de diffusion hylafax'
Objet:	help problem with modem hayes accura

First test is to replace your modem with another one of the same brand.
Though, is has been mentioned that the Hayes Acura may not hang up under
all circumstance, which could be a problem,

Julio

On Tue, 21 Oct 1997, Pierre MEURICE wrote:

> 
> Hello everibody,
> 
> I run hylafax on aix 4.X with a modem Hayes accura 28.8 model 5337,
> and when we send a fax it always answer line is busy.
> 
> Please can you help us.
> Thanck you.
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 21 Oct 1997 16:29:55 -0400
From: Ramana Juvvadi <juvvadi@lekha.org>
To: flexfax@sgi.com
Subject: flexfax: Ghostscript version for hylafax-4.0.i386.rpm
Sender: owner-flexfax@celestial.com


I just realized that there may be a problem with the ghostscript
version for hylafax-4.0. The problem is with the program textfmt.
It expects <font-name>.afm files in /usr/share/ghostscript/fonts.
However ghostscript has a file called Fontmap which provides
the mapping between fontnames and files.

To overcome this problem I created a directory called
/usr/lib/hyla/afm and added symbolic links to the files
in /usr/share/ghostscript/fonts.

I am running with ghostscript-fonts-5.02-1. If you don't have the same
version of ghostscript that I do, there can be a problem with the file
names. I'll have another release with dependency on
ghostscript-fonts-5.02-1 soon.  Meanwhile, if you are having any
problems with ghostscript upgrade to ghostscript-fonts-5.02-1.
It can be found at ftp://ftp.redhat.com/pub/contrib/i386

Ramana

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 12:49:27 +0200
From: Uri Shkolnik <noaml@netvision.net.il>
Reply-To: uri@harmonic.co.il
To: flexfax@sgi.com
Subject: flexfax: Viewing received faxes via http browser
Sender: owner-flexfax@celestial.com

I'm running Hylafax 4p1 on SunOS4.1.4, I want to enabled users to view
faxes with Netscape/Internet Explorer etc. via my Intranet web site.

Does anybody have a working system like this?

Thanks,

Uri

uri@harmonic.co.il

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 15:23:29 +0300
To: uri@harmonic.co.il, flexfax@sgi.com
From: Mario Misic <mario@laus.hr>
Subject: Re: flexfax: Viewing received faxes via http browser
Sender: owner-flexfax@celestial.com

At 12:49 1997.10.22 +0200, Uri Shkolnik wrote:
>I'm running Hylafax 4p1 on SunOS4.1.4, I want to enabled users to view
>faxes with Netscape/Internet Explorer etc. via my Intranet web site.
>
>Does anybody have a working system like this?
>
>Thanks,
>
>Uri
>
>uri@harmonic.co.il
>

Hi!

In config file /var/spool/fax/etc/config.tty* you have a variable
RecvFileMode change it to look like this:

.
.
.
RecvFileMode:		0644
.
.
.

By

M2


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~ Mario Misic              | e-mail: mario@laus.hr       ~
~ CC Computer Consulting   | Tel:    +385 (20) 411-136   ~
~ Janjevska 15             |         +385 (1) 6552-330   ~
~ 20 000 Dubrovnik         | Fax:    +385 (20) 411-136   ~
~ Hrvatska (Croatia)       | URL:    http://www.laus.hr  ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>>>>>>>>>>>> Every dog will have his day ! <<<<<<<<<<<<<<
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 10:22:30 -0500 (CDT)
From: "Edward S. Marshall" <emarshal@xnet.com>
To: uri@harmonic.co.il
cc: flexfax@sgi.com
Subject: Re: flexfax: Viewing received faxes via http browser
X-SpamBait: spam@logic.net
Sender: owner-flexfax@celestial.com

On Wed, 22 Oct 1997, Uri Shkolnik wrote:
> I'm running Hylafax 4p1 on SunOS4.1.4, I want to enabled users to view
> faxes with Netscape/Internet Explorer etc. via my Intranet web site.
> 
> Does anybody have a working system like this?

You could probably rig up something with the Adobe PDF reader plug-in, and
a liberal application of ps2pdf. ;-) Or, convert it to a ReallyBigJPEG(tm),
and eliminate the need for the PDF plugin.

-- 
#!/usr/local/bin/perl --              Edward S. Marshall <emarshal@xnet.com> --
sub g{my$i=index$t,$_[0];($i%5,int$i/5)}sub h{ substr$t,5*$_[1]+$_[0],1}sub n{(
$_[0]+4)%5}$t='encryptabdfghjklmoqsuvwxz';$c='fxmdwbcmagnyubnyquohyhny'; while(
$c=~s/(.)(.)//){($w,$x)=g$1;($y,$z)=g$2;$w==$y&&($p.=h($w,n$x).h($y,n$z))or$x==
$z&&($p.=h(n$w,$x).h(n$y,$z))or($p.=h($y,$x).h($w,$z))}$p=~y/x/ /;print$p,"\n";

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 12:05:43 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: uri@harmonic.co.il
Cc: flexfax@sgi.com
Subject: Re: flexfax: Viewing received faxes via http browser
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Wed, 22 Oct 1997, Uri Shkolnik wrote:

> I'm running Hylafax 4p1 on SunOS4.1.4, I want to enabled users to view
> faxes with Netscape/Internet Explorer etc. via my Intranet web site.
> 
> Does anybody have a working system like this?

YES! Moo-ha-ha! There was a script published called "faxstat.pl" for
exactly this use. I enclose it below. For additional questions, ask
the list and feel free to CC: me.

   #!/usr/local/bin/perl -wT

   =head1 NAME

   faxstat.pl - CGI interface to interrogating HylaFAX server

   =head1 SYNOPSIS

    http://www/cgi-bin/faxstat.pl
    http://www/cgi-bin/faxstat.pl?FAXSERVER=localhostname.net
    http://www/cgi-bin/faxstat.pl?j=2000
    http://www/cgi-bin/faxstat.pl?ps=docq%2Fdoc17777.ps
    http://www/cgi-bin/faxstat.pl?fax=recvq%2Ffax00009.fax

    http://www/cgi-bin/faxstat.pl?d=2000

   =head1 DESCRIPTION

   C<faxstat.pl> implements a web interface to querying a HylaFAX server,
   in a manner similar to L<faxstat(1)>.

   With no HTTP parameters, an HTML page is generated listing faxes in the
   sendq (queued, not yet sent), the doneq (queued and sent), and the
   recvq (faxes received).

   With a C<j=nnnn> parameter, an HTML page is generated containing all the
   job info for job nnnn.

   With a C<ps=xxx> parameter, the script attempts to access the document named xxx
   and serve it as content type application/postscript

   With a C<fax=xxx> parameter, the script attempts to access the document named xxx
   and serve is as content type image/fax

   With a C<d=nnnn> parameter, the script attempts to delete the pending (or
   completed) job nnnn.

   =head2 Other HTTP parameters

   =over 4

   =item FAXSERVER

   FAXSERVER may be used with any of the above to nominate the host which
   the script should attempt to query.  In practice it might be easier to
   edit the code.

   =item debug

   The value of the C<debug> parameter will be passed to the C<debug>
   method of the C<Net::FTP> object used internally.  Note that debugging
   is to STDERR so will probably appear in your server log

   =back

   =head1 BUGS

   There is no way to contact the HylaFAX server as a particular
   (authenticated) user, other than by relying on the www server to do
   authentication and to require no passwords on the hylafax server.
   That sucks.

   Doesn't read the system hyla.conf file for FAXSERVER


   =head1 AUTHOR

   Phil Abercrombie <phil@mcs.vuw.ac.nz>, September 1997

   =cut
   #';# emacs

   # <PJA> 19-Sep-1997
   # Added POD

   use CGI;
   use Config;
   use Net::FTP;

   my $cgi = new CGI;
   $FAXSERVER = $cgi->param("FAXSERVER")||"localhostname.com";
   $HYLAFAX   = 4559;


   eval{&main};
   if ($@)
   {
     my $a=$@;
     print CGI->header,CGI->start_html("Error"), "<P>", join ("<BR>", split "\n", "$a"), CGI->end_html;
     die $a
   }



   sub main
   {
     my $hfax = new Net::FTP $FAXSERVER, Port=>$HYLAFAX
       or die "Can't connect to host $FAXSERVER";
     $hfax->debug($cgi->param("debug"));
     $hfax->login(CGI->remote_user)
       or die "Can't login to host $FAXSERVER";
     $hfax->binary;

     if ($cgi->param("ps"))
     {
       &psinfo($hfax, $cgi->param("ps"))
     }
     elsif ($cgi->param("tiff"))
     {
       &tiffinfo($hfax, $cgi->param("tiff"))
     }
     elsif ($cgi->param("d"))
     {
       &killjob($hfax, $cgi->param("d"))
     }
     elsif ($cgi->param("j"))
     {
       &jobinfo($hfax, $cgi->param("j"))
     }
     else
     {
       &faxstat($hfax)
     }
   }

   sub list ($@)
   {
     my $hfax = shift;
     my $cx = $hfax->list(@_) || die $hfax->message;
     my @results;

     # <PJA> 19-Sep-1997
     # handle continuation lines correctly
     while(<$cx>)
     {
       if($results[$#results]=~ /\\\n$/)
       {
	 $results[$#results].=$_;
       }
       else
       {
	 push(@results,$_);
       }
     }
     $cx->close;
     @results;
   }

   sub list_jobs($@)
   {
     my $hfax = shift;
     map do { my %d; @d{"a".."z","A".."Z"}=split /$;/; \%d }, list $hfax, @_;
   }

   sub faxstat
   {
     my $hfax = shift;
     my $jobfmt = join "$;", map "%$_", "a".."z", "A".."Z";

     print CGI->header;
     print CGI->start_html("Fax System Status");
     print "<H1>Fax System Status</H1>\n";


     my @status = map [split(/:/)], list $hfax, "status";

     $hfax->quot("JOBFMT","\"$jobfmt\"") or die $hfax->message;
     my @send = list_jobs $hfax, "sendq";
     $hfax->quot("JOBFMT","\"$jobfmt\"");
     die $hfax->message unless $hfax->ok;
     my @done = list_jobs $hfax, "doneq";
     $hfax->quot("RCVFMT","\"$jobfmt\"");
     die $hfax->message unless $hfax->ok;
     my @recv = list_jobs $hfax, "recvq";

     print "<H2>Status</H2><TABLE BORDER>\n";
     for(@status)
     {
       for($_->[1])
       {
	 if (/job (\d+)/)
	 {
	   $cgi->param(j=>$1);
	   my $surl = $cgi->self_url;
	   s!job (\d+)!job <A href="$surl">$1</a>!g;
	   $cgi->delete("j");
	 }
       }
       print "<TR>", map( "<TD>$_</TD>", @$_), "</TR>\n";
     }
     print "</TABLE>\n";

     &show_outq( "Jobs in Send Queue", @send );
     &show_outq( "Jobs in Done Queue", @done );
     &show_inq( "Jobs in Receive Queue", @recv );

     print CGI->end_html;
   }

   sub show_outq
   {
     my $title = shift;
     my @jobs = @_;

     if (@jobs)
     {
       print qq{<H2>$title</H2>\n<TABLE BORDER>\n};
       print "<TR>",map "<TH>$_</TH>",
       "Job", "Sender", "To","Dials","Pages","Status";
       print "</TR>\n";
       for(@jobs)
       {
	 my $k;
	 for $k (keys %$_){$_->{$k} ||= "&nbsp;"};

	 $_->{"e"} =~ s/^\+644/+64&nbsp;4&nbsp;/;
	 $_->{"s"} =~ s/ /&nbsp;/g;

	 $cgi->param(j=>$_->{j});
	 my $self = $cgi->self_url;
	 print <<JOB;
   <TR>
   <TD><A HREF="$self">$_->{j}</a></TD>
   <TD><A HREF="mailto:$_->{M}">$_->{M}</a></TD>
   <TD><TT>$_->{e}</TT></TD>
   <TD>$_->{D}</TD>
   <TD>$_->{P}</TD>
   <TD>$_->{"s"}</TD>
   </TR>
   JOB
     ;
       }
       print "</TABLE>\n";
     }
   }

   sub show_inq
   {
     my $title = shift;
     my @jobs = @_;
     my $self = $cgi->self_url;

     if (@jobs)
     {
       print qq{<H2>$title</H2>\n<TABLE BORDER>\n};
       print "<TR>",map "<TH>$_</TH>",
       "Filename", "Sender/TSI", "Pages","Received@";
       print "</TR>\n";
       for(@jobs)
       {
	 $cgi->param(tiff=>"recvq/$_->{f}");
	 my $myself = $cgi->self_url;

	 print <<JOB;
   <TR>
   <TD><A HREF="$myself">$_->{f}</a></TD>
   <TD><TT>$_->{"s"}</TT></TD>
   <TD>$_->{p}$_->{z}</TD>
   <TD>$_->{t}</TD>
   </TR>
   JOB
     ;
       }
       print "</TABLE>\n";
     }
   }

   sub jobinfo
   {
     my $hfax = shift;
     my $job = shift;
     my $cx;
     my $self = CGI->script_name;

     print CGI->header;
     print CGI->start_html("Fax Job Info");

     print "<H1>Job Info for job $job</H1>\n";

     $cx = $hfax->retr( "sendq/q$job" ) || $hfax->retr( "doneq/q$job" ) || die $hfax->message;
     my @job;
     while(<$cx>)
     {
       if(@job && $job[$#job] =~ s/\\$/<BR>/)
       {
	 $job[$#job] .= $_;
       }
       else
       {
	 push(@job,$_);
       }    
     }

     die "No job info available" unless @job;
     $cx->close or die "Data connection error";

     $cgi->param(d=>$job);
     my $myself=$cgi->self_url;
     print qq{<A HREF="$myself">Delete Me</A>\n};
     $cgi->delete('d');

     print "<TABLE BORDER>\n";
     for (@job)
     {
       chomp;
       $cgi->delete("ps","tiff");
       my ($key,$val)=split(/:/,$_,2);
       $val = localtime($val) if $key eq 'tts' || $key =~ /time$/;
       if ($key =~ /postscript|tiff|fax/)
       {
	 my $param = ($& eq "postscript")?"ps":"tiff";
	 my ($ref) = ($val =~ m!(docq/\S+)!);
	 $cgi->param($param,$ref);
	 my $myself = $cgi->self_url;
	 $val =~ s!docq/\S+!<a href="$myself">$&</a>!;
       }

       print "<TR><TD>$key</TD><TD>$val</TD></TR>\n";
     }
     print "</TABLE>\n";

     print CGI->end_html;
   }

   sub psinfo
   {
     my $hfax = shift;
     my $psfile = shift;

     $cx = $hfax->retr( $psfile ) || die $hfax->message;
     my @job=<$cx>;
     $cx->close or die "Data connection error";

     print CGI->header( "application/postscript" );
     print @job;
   }

   sub tiffinfo
   {
     my $hfax = shift;
     my $tiff = shift;

     $cx = $hfax->retr( $tiff ) || die $hfax->message;
     my @job=<$cx>;
     $cx->close or die "Data connection error";

     print CGI->header( "image/fax" );
     print @job;
   }

   sub killjob
   {
     my $hfax = shift;
     my $job = shift;
     $hfax->quot( "JSUSP", $job );
     my ($mess,$ok)=($hfax->message, $hfax->ok);
     $hfax->quot( "JDELE", $job );
     die "$mess\n".$hfax->message unless $hfax->ok;

     print CGI->redirect( $cgi->self_url );
   }


-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNE4kUT/+ItycgIJRAQGoygQAl2bkHLmUV4dBO+acXprYaQO98NUvW99D
aLzIDQ55df30EUnlUlty8tiplh5QY86KLN97xAsvVk+LupAsqc8hKjWBwYdbgNkp
uj5y/Qsp1a6flWcxOtzOXhAE6lyEH0/uiIVYLSOk/mlRQCVfqU+5JYtez2B3MJEY
By/mqq+tLtk=
=cHVV
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 12:12:40 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Mario Misic <mario@laus.hr>
Cc: Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: Viewing received faxes via http browser
Sender: owner-flexfax@celestial.com

On Wed, 22 Oct 1997, Mario Misic wrote:

> At 12:49 1997.10.22 +0200, Uri Shkolnik wrote:
> >I'm running Hylafax 4p1 on SunOS4.1.4, I want to enabled users to view
> >faxes with Netscape/Internet Explorer etc. via my Intranet web site.
> >
> >Does anybody have a working system like this?
> >Thanks,
> >Uri
> >
> >uri@harmonic.co.il
> >
> 
> Hi!
> 
> In config file /var/spool/fax/etc/config.tty* you have a variable
> RecvFileMode change it to look like this:
> 
> RecvFileMode:		0644
> .

This is not sufficient. On further thought, I remember a couple of other
things I did to get it to work.

First, the ".tif" extension on received extensions is extremely confusing.
It is worth modifying the faxrcvd to use the extension ".fax" instead.

Second, users must have a tool capable of viewing either delivered
Postscript files or the raw tiffg3 files. I recommend "faxview" at
ftp://ftp.sgi.com/sgi/fax/contrib/ for UNIX based X servers.

Third, they have to teach their viewer to use the correct tool to view
".fax" files.

The "faxstat.pl" CGI script I just published was *not* written by me:
the author, Phil Abercrombie, is phil@mcs.vuw.ac.nz and deserves credit.

An alternative approach would be to translate the tiffg3 files into a
set of plain tiff files, one per page, and let people view *those*. This
is left as an exercise for the CGI programmer....


			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 18:49:55 +0200
From: Uri Shkolnik <noaml@netvision.net.il>
Reply-To: uri@harmonic.co.il
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
CC: uri@harmonic.co.il, flexfax@sgi.com
Subject: Re: flexfax: Viewing received faxes via http browser
Sender: owner-flexfax@celestial.com



Nico Garcia wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
>
> On Wed, 22 Oct 1997, Uri Shkolnik wrote:
>
> > I'm running Hylafax 4p1 on SunOS4.1.4, I want to enabled users to view
> > faxes with Netscape/Internet Explorer etc. via my Intranet web site.
> >
> > Does anybody have a working system like this?
>
> YES! Moo-ha-ha! There was a script published called "faxstat.pl" for
> exactly this use. I enclose it below. For additional questions, ask
> the list and feel free to CC: me.

Hello,

Thanks for the replay.

I had tried to cut the perl script from your Email and copy it into my cgi-bin, but I have some errors.

this is the beginning of the script:

#!/usr/local/bin/perl -wT
  #';# emacs

   # <PJA> 19-Sep-1997
   # Added POD

   use CGI;
   use Config;
   use Net::FTP;

   my $cgi = new CGI;
   $FAXSERVER = $cgi->param("FAXSERVER")||"fax.harmonic.co.il";
   $HYLAFAX   = 4559;



There is an server error, I tried to run it directly: --

xxx{root}: sh faxstat.pl
faxstat.pl: use: not found
faxstat.pl: use: not found
faxstat.pl: use: not found
faxstat.pl: my: not found
faxstat.pl: syntax error at line 12: `(' unexpected


What are those errors?
Could you send me the script as attacment?

Thanks & Regards,

Uri

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 09:22:51 +0300
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
From: Mario Misic <mario@laus.hr>
Subject: Re: flexfax: Viewing received faxes via http browser
Cc: Nico Garcia <flexfax@sgi.com>
Sender: owner-flexfax@celestial.com

At 12:12 1997.10.22 -0400, Nico Garcia wrote:
>-----BEGIN PGP SIGNED MESSAGE-----
>
>On Wed, 22 Oct 1997, Mario Misic wrote:
>
>> At 12:49 1997.10.22 +0200, Uri Shkolnik wrote:
>> >I'm running Hylafax 4p1 on SunOS4.1.4, I want to enabled users to view
>> >faxes with Netscape/Internet Explorer etc. via my Intranet web site.
>> >
>> >Does anybody have a working system like this?
>> >
>> >Thanks,
>> >
>> >Uri
>> >
>> >uri@harmonic.co.il
>> >
>> 
>> Hi!
>> 
>> In config file /var/spool/fax/etc/config.tty* you have a variable
>> RecvFileMode change it to look like this:
>> 
>> .
>> .
>> .
>> RecvFileMode:		0644
>> .
>> .
>> .
>
>This is not sufficient. On further thought, I remember a couple of other
>things I did to get it to work.
>
>First, the ".tif" extension on received extensions is extremely confusing.
>It is worth modifying the faxrcvd to use the extension ".fax" instead.
>
>Second, users must have a tool capable of viewing either delivered
>Postscript files or the raw tiffg3 files. I recommend "faxview" at
>ftp://ftp.sgi.com/sgi/fax/contrib/ for UNIX based X servers.
>
>Third, they have to teach their viewer to use the correct tool to view
>".fax" files.
>
>The "faxstat.pl" CGI script I just published was *not* written by me:
>the author, Phil Abercrombie, is phil@mcs.vuw.ac.nz and deserves credit.
>
>An alternative approach would be to translate the tiffg3 files into a
>set of plain tiff files, one per page, and let people view *those*. This
>is left as an exercise for the CGI programmer....
>
>
>

Hi

Yes of course it is possible, but you already have a plug-in for Netscape
for viewing a faxes, all you have to do is to modify config.tty* file, but
security is other thing.

By
M2


-----
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~ Mario Misic              | e-mail: mario@laus.hr       ~
~ CC Computer Consulting   | Tel:    +385 (20) 411-136   ~
~ Janjevska 15             |         +385 (1) 6552-330   ~
~ 20 000 Dubrovnik         | Fax:    +385 (20) 411-136   ~
~ Hrvatska (Croatia)       | URL:    http://www.laus.hr  ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>>>>>>>>>>>> Every dog will have his day ! <<<<<<<<<<<<<<
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 09:09:01 +0200
From: Ulrich Eckhardt <uli@transcom.de>
Organization: Transcom
To: HylaFax Mailing liste <flexfax@sgi.com>
CC: uri@harmonic.co.il
Subject: Re: flexfax: Viewing received faxes via http browser
Sender: owner-flexfax@celestial.com

Uri Shkolnik wrote:
> 
> I'm running Hylafax 4p1 on SunOS4.1.4, I want to enabled users to view
> faxes with Netscape/Internet Explorer etc. via my Intranet web site.
> 
> Does anybody have a working system like this?
> 
> Thanks,
> 
> Uri
> 
> uri@harmonic.co.il

Hi,

i have written here a perlscript which displays faxes via web and
allows users to delete faxes (all users which have access to this
page). It requires a running web server on the fax-host and 
RecvFileMode should be set to 0666. If someone is interested i have
put the script on our ftp server ftp://ftp.transcom.de/pub/webview.

You find here show.pl which is the script and fax.gif which is a small
image. Put show.pl in the cgi-bin directory and fax.gif in the directory
docroot/graphics. Ckeck the first lines of the script, because the
pathes to tiffcp and the png-programms may differ.

Uli
-- 
Ulrich Eckhardt		                   mailto:uli@transcom.de
http://people.frankfurt.netsurf.de/uli
Truly great madness can not be achieved without significant
intelligence. (Henrik Tikkanen)

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 11:20:21 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: "Edward S. Marshall" <emarshal@xnet.com>
Cc: Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: Viewing received faxes via http browser
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Wed, 22 Oct 1997, Edward S. Marshall wrote:

> On Wed, 22 Oct 1997, Uri Shkolnik wrote:
> > I'm running Hylafax 4p1 on SunOS4.1.4, I want to enabled users to view
> > faxes with Netscape/Internet Explorer etc. via my Intranet web site.
> > 
> > Does anybody have a working system like this?
> 
> You could probably rig up something with the Adobe PDF reader plug-in, and
> a liberal application of ps2pdf. ;-) Or, convert it to a ReallyBigJPEG(tm),
> and eliminate the need for the PDF plugin.

*OUCH*. What would be ideal is a reader (such as viewfax, which I use)
to view the raw tiffg3 files (there is an OK demo one available for
Win95, I believe called acdsee). Also workable is translating the
tiffg3 into a set of multiple single page TIFF files for
viewing. Doing tiffg3->Postscript->JPEG is just sucking down CPU and
disk space and bandwidth in a *serious* kludge.

Hmm. Come to think of it: Does anyone have the scripting in place to
modify faxrcvd to automagically generate TIFF files, one file/page of
fax and appropriately numbered? This might be a good option for
handling Web based FAX service....


			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNE9rKj/+ItycgIJRAQHC/wP9HV5bqTZy8tSNZ9v5TYVSiyO98Gs5yYWz
G6OcCuqSzDwQjTSGfiDLTVyUBAq1cMPLDXRbFMzcHDWGhs4VdlFqrrf2/gasRJ0L
OG/6EaEWWzQTLRYgXys0Ivoir2Pzsb429r1fCCGHc0f1HT6o682h+fwHq+CG5oCX
KCcufLdCzoc=
=4vS2
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 11:44:01 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Mario Misic <mario@laus.hr>
Cc: Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: Viewing received faxes via http browser
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Thu, 23 Oct 1997, Mario Misic wrote:

> >An alternative approach would be to translate the tiffg3 files into a
> >set of plain tiff files, one per page, and let people view *those*. This
> >is left as an exercise for the CGI programmer....
> >
> >
> >
> 
> Hi
> 
> Yes of course it is possible, but you already have a plug-in for Netscape
> for viewing a faxes, all you have to do is to modify config.tty* file, but
> security is other thing.

Which plug-in is this? Is it available for SunOS, Linux, Mac, Win95,
WinNT, and all the other platforms?

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNE9wuz/+ItycgIJRAQEZrAP/XbPxkokl6I9bg0ZGf+BS9ZAfeBVdZBP1
iwtigwmv0OtTg1QYR84hiQZJCrD8cNrfWk4OANDzVezkrXsSpgTUh03rAmtsCGC2
8M57WAufyBKwTiS1Wd7JDXYa5dWyZQ0wEDbfWnoEVBuH6Oz1b4jClrOX7iDcKWYW
qd+2oFLSMLc=
=GDWh
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 11:51:11 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Nico Garcia <flexfax@sgi.com>
Subject: flexfax: Re: Viewing received faxes via http browser
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Thu, 23 Oct 1997, Hong Kong Observatory wrote:

> Hi Nico,
> 
> I just extracted your faxstat.pl sugested to Uri Shkolnik. But when
> I tried to run the script, system complained cannot found Net/FTP
> package. Please tell where to find the required package.
> 
> I am using Perl5.004 under Linux 2.0.28.
> 
> Thanks !


I've gotten other requests, so I'm posting this answer. You'll find
this and other perl packages at:

	http://www.perl.com/CPAN-local/CPAN.html

Net::FTP is part of the "libnet" bundle, which is fairly easy to install.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu


-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNE9yaT/+ItycgIJRAQFtuQQAhf0GMMD/XdRSTuSiKmrgQO1dJObnUU+f
MnpD1syFSML6+OGxIcSuITF35MjLIZX+EB2zg3LLaoFFsfgMDi19CnI40wUsWoOp
guo1VShJOobGutWTquEbbe9kPOocPzP/DMhviXMOxxtByM01fuF9+lFHcfvALPYS
Lsc0Ynx/e8o=
=tXht
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 12:22:57 -0400 (EDT)
From: Rob Newberry <rob@eats.com>
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
cc: "Edward S. Marshall" <emarshal@xnet.com>,
        hylaFAX mailing list <flexfax@sgi.com>
Subject: Re: flexfax: Viewing received faxes via http browser
Sender: owner-flexfax@celestial.com


> *OUCH*. What would be ideal is a reader (such as viewfax, which I use)
> to view the raw tiffg3 files (there is an OK demo one available for
> Win95, I believe called acdsee). Also workable is translating the
> tiffg3 into a set of multiple single page TIFF files for
> viewing. Doing tiffg3->Postscript->JPEG is just sucking down CPU and
> disk space and bandwidth in a *serious* kludge.
> 
> Hmm. Come to think of it: Does anyone have the scripting in place to
> modify faxrcvd to automagically generate TIFF files, one file/page of
> fax and appropriately numbered? This might be a good option for
> handling Web based FAX service....

It's not web-based, but I've done something similar.

I wrote a FAX viewer for the Mac -- it reads raw TIF G3 (heck, it uses
LIBTIFF -- thanks, Sam).  It's part of my MacFlex distribution, the
Macintosh client for HylaFAX.  Find out more at
http://www.eats.com/MacFlex.

Anyway, I've modified faxrcvd to automatically binhex the received TIFF
file and email it to my wife as an attachment.  They appear on her Mac
desktop, and she uses my viewer to look at them.

But I ALSO have the fax recvq shared via netatalk so my Macs can mount it
directly.  The faxrcvd script also sends me a message that a FAX has
arrived; when I want to, I just mount the FAX spool and view the FAXes
directly (again, using my viewer).

It should be possible to do something similar for Windows using SAMBA.
I've been working on a port of my Mac viewer to Windows (for fun mainly).

It's not web-based, but it certainly solves my needs pretty elgantly.

(the only thing cooler would be to automatically rename the files so that
I can tell who they are from -- or to use the HylaFAX protocol to access
them, rather than the file system).

Rob

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 22:27:41 +0200
From: Stefan Wintermeyer <sw@swr.de>
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
CC: "Edward S. Marshall" <emarshal@xnet.com>, Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: Viewing received faxes via http browser
Sender: owner-flexfax@celestial.com

Nico Garcia wrote:
> Win95, I believe called acdsee). Also workable is translating the
> tiffg3 into a set of multiple single page TIFF files for
> viewing. Doing tiffg3->Postscript->JPEG is just sucking down CPU and
> disk space and bandwidth in a *serious* kludge.

Not a good idea. We are converting them to GIFs. This is much smaller
because faxes do not have many colours.
-- 
Stefan Wintermeyer
SWR Reisedienst GmbH
Tel. +49 2631 9600             http://www.swr.de
Fax  +49 2631 960-111          mailto:info@swr.de

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 17:09:45 -0500 (CDT)
From: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
To: Stefan Wintermeyer <sw@swr.de>
cc: Nico Garcia <raoul@cirl.meei.harvard.edu>,
        "Edward S. Marshall" <emarshal@XNET.COM>,
        Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: Viewing received faxes via http browser
Sender: owner-flexfax@celestial.com

Hi Stefan,

I need to systematically send TIFF G3 files over the modem line, I don't
know much about image formats, what would be the smallest format that I
could use (they are faxes at low resolution) and what would be the best
compression program to use if any?

Would you happen to know where to find them? I would need to be able to go
back to TIFF G3 or to some format that hylafax can read,

Thanks in advance for your help,

Julio

On Thu, 23 Oct 1997, Stefan Wintermeyer wrote:

> Nico Garcia wrote:
> > Win95, I believe called acdsee). Also workable is translating the
> > tiffg3 into a set of multiple single page TIFF files for
> > viewing. Doing tiffg3->Postscript->JPEG is just sucking down CPU and
> > disk space and bandwidth in a *serious* kludge.
> 
> Not a good idea. We are converting them to GIFs. This is much smaller
> because faxes do not have many colours.
> -- 
> Stefan Wintermeyer
> SWR Reisedienst GmbH
> Tel. +49 2631 9600             http://www.swr.de
> Fax  +49 2631 960-111          mailto:info@swr.de
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 19:13:07 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
Cc: Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: Viewing received faxes via http browser
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Thu, 23 Oct 1997, Prof. Julio C. Spinelli wrote:

> Hi Stefan,
> 
> I need to systematically send TIFF G3 files over the modem line, I don't
> know much about image formats, what would be the smallest format that I
> could use (they are faxes at low resolution) and what would be the best
> compression program to use if any?

tiffg3 is actually pretty damn compressed. I would leave it alone as
much as possible.

> Would you happen to know where to find them? I would need to be able to go
> back to TIFF G3 or to some format that hylafax can read,

The TIFF libraries include "tiffcp" and "fax2tiff", which will very
effectively convert tiffg3 to the more standard TIFF, one per
page. You can then turn these into arbitrary image formats. What I've
never set up is getting them into organized, numbered files with
the right suffix numbers. For example, a 12 page fax should be:

	fax00001.fax  ->	fax0001.tif.01
				fax0001.tif.02
				...
				fax0001.tif.09
				fax0001.tif.10
				fax0001.tif.11
				fax0001.tif.12

so that they are listed in correct alphabetical order by most browsers.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNE/Z+D/+ItycgIJRAQGpYAP/ZC9EZueTtingDAjLB8qcv6fWqA3UHGnH
kHTg2HAujjvOF0wIEcsXScXv+sdryDrTXLUTTWG1VsPBLAbHhVpVb1Ayp+jtm4Jy
ZNEjPHNEhzptBzNmQhCVNArYr24AYWN2h8MIer/mTa0nflrh/X3Bdcb1WTbGUS1s
jLWHn508eaw=
=EnP4
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 24 Oct 1997 09:15:22 +0300
To: flexfax@sgi.com
From: Mario Misic <mario@laus.hr>
Subject: Re: flexfax: Viewing received faxes via http browser
Sender: owner-flexfax@celestial.com

At 12:12 1997.10.22 -0400, Nico Garcia wrote:
>-----BEGIN PGP SIGNED MESSAGE-----
>
>On Wed, 22 Oct 1997, Mario Misic wrote:
>
>> At 12:49 1997.10.22 +0200, Uri Shkolnik wrote:
>> >I'm running Hylafax 4p1 on SunOS4.1.4, I want to enabled users to view
>> >faxes with Netscape/Internet Explorer etc. via my Intranet web site.
>> >
>> >Does anybody have a working system like this?
>> >
>> >Thanks,
>> >
>> >Uri
>> >
>> >uri@harmonic.co.il
>> >
>> 
>> Hi!
>> 
>> In config file /var/spool/fax/etc/config.tty* you have a variable
>> RecvFileMode change it to look like this:
>> 
>> .
>> .
>> .
>> RecvFileMode:		0644
>> .
>> .
>> .
>
>This is not sufficient. On further thought, I remember a couple of other
>things I did to get it to work.
>
>First, the ".tif" extension on received extensions is extremely confusing.
>It is worth modifying the faxrcvd to use the extension ".fax" instead.
>
>Second, users must have a tool capable of viewing either delivered
>Postscript files or the raw tiffg3 files. I recommend "faxview" at
>ftp://ftp.sgi.com/sgi/fax/contrib/ for UNIX based X servers.
>
>Third, they have to teach their viewer to use the correct tool to view
>".fax" files.
>
>The "faxstat.pl" CGI script I just published was *not* written by me:
>the author, Phil Abercrombie, is phil@mcs.vuw.ac.nz and deserves credit.
>
>An alternative approach would be to translate the tiffg3 files into a
>set of plain tiff files, one per page, and let people view *those*. This
>is left as an exercise for the CGI programmer....
>
>
>

Hi

Yes of course it is possible, but you already have a plug-in for Netscape
for viewing a faxes, all you have to do is to modify config.tty* file,
export file system and that is all !

By
M2


-----
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~ Mario Misic              | e-mail: mario@laus.hr       ~
~ CC Computer Consulting   | Tel:    +385 (20) 411-136   ~
~ Janjevska 15             |         +385 (1) 6552-330   ~
~ 20 000 Dubrovnik         | Fax:    +385 (20) 411-136   ~
~ Hrvatska (Croatia)       | URL:    http://www.laus.hr  ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>>>>>>>>>>>> Every dog will have his day ! <<<<<<<<<<<<<<
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 24 Oct 1997 09:28:23 +0300
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
From: Mario Misic <mario@laus.hr>
Subject: Re: flexfax: Viewing received faxes via http browser
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

At 11:44 1997.10.23 -0400, Nico Garcia wrote:
>-----BEGIN PGP SIGNED MESSAGE-----
>
>On Thu, 23 Oct 1997, Mario Misic wrote:
>
>> >An alternative approach would be to translate the tiffg3 files into a
>> >set of plain tiff files, one per page, and let people view *those*. This
>> >is left as an exercise for the CGI programmer....
>> >
>> >
>> >
>> 
>> Hi
>> 
>> Yes of course it is possible, but you already have a plug-in for Netscape
>> for viewing a faxes, all you have to do is to modify config.tty* file, but
>> security is other thing.
>
>Which plug-in is this? Is it available for SunOS, Linux, Mac, Win95,
>WinNT, and all the other platforms?
>
Hi

It is available for Win95/NT and name of plug-in is CPCVPlug.exe and it is
available from netscape wwwsite (www.netscape.com).

Last time I didn't find plug-in for UNIX (we are on AIX platform)!

Try it it is easy to use and easy to install !!

By

M2
-----
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~ Mario Misic              | e-mail: mario@laus.hr       ~
~ CC Computer Consulting   | Tel:    +385 (20) 411-136   ~
~ Janjevska 15             |         +385 (1) 6552-330   ~
~ 20 000 Dubrovnik         | Fax:    +385 (20) 411-136   ~
~ Hrvatska (Croatia)       | URL:    http://www.laus.hr  ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>>>>>>>>>>>> Every dog will have his day ! <<<<<<<<<<<<<<
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 24 Oct 97 10:02:32 BST
From: Neil Padgen <nrp@chernikeeff.co.uk>
To: flexfax@sgi.com
Subject: Re: flexfax: Viewing received faxes via http browser
Newsgroups: chernikeeff.mail.hylafax
Organization: Chernikeeff
Sender: owner-flexfax@celestial.com

In article <344DDA36.378AC735@netvision.net.il> you write:
>I'm running Hylafax 4p1 on SunOS4.1.4, I want to enabled users to view
>faxes with Netscape/Internet Explorer etc. via my Intranet web site.
>
>Does anybody have a working system like this?
>
>Thanks,
>
>Uri
>
>uri@harmonic.co.il

I've written a set of perl CGI scripts to do this.  The faxrecv script
uses the ImageMagick convert utility to convert the fax to JPEG,
generating multiple images for a multiple-page fax.  The CGI talks
to the fax server using FTP.pm and displays the faxes within the
browser (no plugin or external viewer needed).

If anyone would like a copy let me know.

-- Neil

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 24 Oct 97 14:57:42 BST
From: Neil Padgen <nrp@chernikeeff.co.uk>
To: flexfax@sgi.com
Subject: Re: flexfax: Viewing received faxes via http browser
Newsgroups: chernikeeff.mail.hylafax
Organization: Chernikeeff
Sender: owner-flexfax@celestial.com

In article <9710240902.AA19234@chart.chernikeeff.co.uk> I wrote:
>In article <344DDA36.378AC735@netvision.net.il> you write:
>>I'm running Hylafax 4p1 on SunOS4.1.4, I want to enabled users to view
>>faxes with Netscape/Internet Explorer etc. via my Intranet web site.
>>
>>Does anybody have a working system like this?
>>
>>Thanks,
>>
>>Uri
>>
>>uri@harmonic.co.il
>
>I've written a set of perl CGI scripts to do this.  The faxrecv script
>uses the ImageMagick convert utility to convert the fax to JPEG,
>generating multiple images for a multiple-page fax.  The CGI talks
>to the fax server using FTP.pm and displays the faxes within the
>browser (no plugin or external viewer needed).
>
>If anyone would like a copy let me know.
>
>-- Neil

I know it's bad form to follow up to one's own message, but I've had
quite a few requests for this since posting, so I'm sending it to the
list.

Sorry for having to uuencode - there's a Makefile included and I don't
want tabs to get lost.

-- Neil


Attachment Converted: "C:\PROGRAM FILES\EUDORA\Attach\hylafax-www.tar.gz"

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 24 Oct 1997 09:54:38 -0500
From: Lachlan Dunlop <lach@lach.net>
To: Mario Misic <mario@laus.hr>
CC: Nico Garcia <raoul@cirl.meei.harvard.edu>, flexfax@sgi.com
Subject: Re: flexfax: Viewing received faxes via http browser
Sender: owner-flexfax@celestial.com

I just asked the floks that make the CVCVP plug-in.  They have no intensions of
porting to the Macintosh or Unix at time. :((

Lach

Mario Misic wrote:

> At 11:44 1997.10.23 -0400, Nico Garcia wrote:
> >-----BEGIN PGP SIGNED MESSAGE-----
> >
> >On Thu, 23 Oct 1997, Mario Misic wrote:
> >
> >> >An alternative approach would be to translate the tiffg3 files into a
> >> >set of plain tiff files, one per page, and let people view *those*. This
> >> >is left as an exercise for the CGI programmer....
> >> >
> >> >
> >> >
> >>
> >> Hi
> >>
> >> Yes of course it is possible, but you already have a plug-in for Netscape
> >> for viewing a faxes, all you have to do is to modify config.tty* file, but
> >> security is other thing.
> >
> >Which plug-in is this? Is it available for SunOS, Linux, Mac, Win95,
> >WinNT, and all the other platforms?
> >
> Hi
>
> It is available for Win95/NT and name of plug-in is CPCVPlug.exe and it is
> available from netscape wwwsite (www.netscape.com).
>
> Last time I didn't find plug-in for UNIX (we are on AIX platform)!
>
> Try it it is easy to use and easy to install !!
>
> By
>
> M2
> -----
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> ~ Mario Misic              | e-mail: mario@laus.hr       ~
> ~ CC Computer Consulting   | Tel:    +385 (20) 411-136   ~
> ~ Janjevska 15             |         +385 (1) 6552-330   ~
> ~ 20 000 Dubrovnik         | Fax:    +385 (20) 411-136   ~
> ~ Hrvatska (Croatia)       | URL:    http://www.laus.hr  ~
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> >>>>>>>>>>>>> Every dog will have his day ! <<<<<<<<<<<<<<
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 24 Oct 1997 11:57:05 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Mario Misic <mario@laus.hr>
Cc: flexfax@sgi.com
Subject: Re: flexfax: Viewing received faxes via http browser
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Fri, 24 Oct 1997, Mario Misic wrote:

> It is available for Win95/NT and name of plug-in is CPCVPlug.exe and it is
> available from netscape wwwsite (www.netscape.com).
> 
> Last time I didn't find plug-in for UNIX (we are on AIX platform)!
> 
> Try it it is easy to use and easy to install !!

OK. For SunOS and Linux (and I don't know what else), the viewfax
utility at ftp://ftp.sgi.com/sgi/fax/contrib/ works just fine as an
add-on viewer, easily summoned as a separate application by the
web browser.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNFDFRj/+ItycgIJRAQErLAP/eZKb9BRbxHOZbpr/kuUvDmpPXNfUr8o/
mfyr0KPtdmPRXZBMp3yuVB7mxQrZnDBlJGE55MPTKMo26i25VMLmFF9HNfHG8BlR
8SYaUFz/lAGPpL9zG9g0+H0uN7YDJ8HSm5KC7MhzbyMJf1Ei37lCx5CprNRiC5vq
IF6eLHBWNu8=
=yEeF
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 24 Oct 1997 12:00:47 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Mario Misic <mario@laus.hr>
Cc: flexfax@sgi.com
Subject: Re: flexfax: Viewing received faxes via http browser
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Fri, 24 Oct 1997, Mario Misic wrote:

> >An alternative approach would be to translate the tiffg3 files into a
> >set of plain tiff files, one per page, and let people view *those*. This
> >is left as an exercise for the CGI programmer....


> Yes of course it is possible, but you already have a plug-in for Netscape
> for viewing a faxes, all you have to do is to modify config.tty* file,
> export file system and that is all !

You have a Win95 plug-in for Netscape. Some of use Lynx: Some of us
use Internet Explorer: Some of us use SunOS, or Linux, etc.

What you describe is quite useful, but it is not a *general* solution.
It also replaces your default *.tif viewer with a different one, just
for viewing the faxes named *.tif. I consider it helpful to name the
files *.fax instead, so the viewers and tools like ImagaMagick can
distinguish it from the more "normal" TIFF formats which they know
about.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNFDGHj/+ItycgIJRAQFEmwP+LF1HPF3jFv/+Dmkr50IbZbrbzWHuBpwX
WsQMy0iut5syPWmDG1BxNxum7biwa+7+ARLmEqJCRJV0ngvcIcIbNsvSatpEpSY5
XD5YR3GWfiea+bTdO6380KD+xAGWZwl5XTmd8tOs0grpiUeopxUKcZGy5X3cEU99
JL0OX2nNyH0=
=UzZL
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Sat, 25 Oct 1997 08:37:28 +0300
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
From: Mario Misic <mario@laus.hr>
Subject: Re: flexfax: Viewing received faxes via http browser
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

At 12:00 1997.10.24 -0400, Nico Garcia wrote:
>-----BEGIN PGP SIGNED MESSAGE-----
>
>On Fri, 24 Oct 1997, Mario Misic wrote:
>
>> >An alternative approach would be to translate the tiffg3 files into a
>> >set of plain tiff files, one per page, and let people view *those*. This
>> >is left as an exercise for the CGI programmer....
>
>
>> Yes of course it is possible, but you already have a plug-in for Netscape
>> for viewing a faxes, all you have to do is to modify config.tty* file,
>> export file system and that is all !
>
>You have a Win95 plug-in for Netscape. Some of use Lynx: Some of us
>use Internet Explorer: Some of us use SunOS, or Linux, etc.
>
>What you describe is quite useful, but it is not a *general* solution.
>It also replaces your default *.tif viewer with a different one, just
>for viewing the faxes named *.tif. I consider it helpful to name the
>files *.fax instead, so the viewers and tools like ImagaMagick can
>distinguish it from the more "normal" TIFF formats which they know
>about.

Hi again!

Yes of course, for those who have an intranet app like me that is quite
useful, but the others who use another fax viewer they have to replace
extension fax to tiff or to something else, I agree.

By
M2
-----
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~ Mario Misic              | e-mail: mario@laus.hr       ~
~ CC Computer Consulting   | Tel:    +385 (20) 411-136   ~
~ Janjevska 15             |         +385 (1) 6552-330   ~
~ 20 000 Dubrovnik         | Fax:    +385 (20) 411-136   ~
~ Hrvatska (Croatia)       | URL:    http://www.laus.hr  ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>>>>>>>>>>>> Every dog will have his day ! <<<<<<<<<<<<<<
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: Viewing received faxes via http browser
To: nrp@chernikeeff.co.uk (Neil Padgen)
Date: Sat, 25 Oct 1997 14:56:41 +0100 (BST)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> I've written a set of perl CGI scripts to do this.  The faxrecv script
> uses the ImageMagick convert utility to convert the fax to JPEG,

Ouch.  JPEG is the WRONG format for faxes.  Of the more or less universally
supported browser formats, only XBM and GIF are appropriate for material
with fine monochrome detail.  JPEG is optimised for continous tone images,
where there are few high spatial frequency components.  It produces all
sorts of artefacts on images which are mainly black and white lines.

XBM is, of course, too inefficient for fax sized images.

> generating multiple images for a multiple-page fax.  The CGI talks
> to the fax server using FTP.pm and displays the faxes within the
> browser (no plugin or external viewer needed).

I would have thought that the PBM tools would have ported to Windows
easily, but generally it would be easier to do everything Unix side.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: Viewing received faxes via http browser
To: raoul@cirl.meei.harvard.edu (Nico Garcia)
Date: Sat, 25 Oct 1997 15:18:07 +0100 (BST)
Cc: mario@laus.hr, flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> 
> First, the ".tif" extension on received extensions is extremely confusing.
> It is worth modifying the faxrcvd to use the extension ".fax" instead.

They are TIFF files.

> Second, users must have a tool capable of viewing either delivered
> Postscript files or the raw tiffg3 files. I recommend "faxview" at
> ftp://ftp.sgi.com/sgi/fax/contrib/ for UNIX based X servers.

For Widows NT 4.0, the imaging tool that is bundled will display
the group 3 encoded TIFF files from Hylafax.  With the junk fax I tried
it on, it showed the correct aspect ratio++, although it is possible that
the junk faxer used fine mode.

> An alternative approach would be to translate the tiffg3 files into a
> set of plain tiff files, one per page, and let people view *those*. This
> is left as an exercise for the CGI programmer....

cp is a good utility for this; tiffg3 files *are* TIFF files, although not
all TIFF files are tiffg3.  The problem is that people don't realise that
TIFF is a way of supplying parameters for multiple graphical file formats,
rather more than a graphic format itself.  It can code for raw bit maps,
LZW palletised images (similar to GIF), images coded according to group 3
fax rules, and, I think, even JPEG encoded images.  Unless, by plain
TIFF, you mean the uncompressed bit map with square pixels, you need to
specify what is allowed in your definition of plain tiff.

++ Checking the TIFF spec, it appears that XResolution (0x7x) and YResolution
(0x82) are mandatory tags, so a viewer should always have the information
to correct the aspect ratio, for a valid file, although I can't quickly
check that the files are valid.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 27 Oct 1997 11:07:16 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: David Woolley <david@djwhome.demon.co.uk>
Cc: Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: Viewing received faxes via http browser
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Sat, 25 Oct 1997, David Woolley wrote:

> > 
> > First, the ".tif" extension on received extensions is extremely confusing.
> > It is worth modifying the faxrcvd to use the extension ".fax" instead.
> 
> They are TIFF files.

Fair enough. But many TIFF viewers can't handle them properly: for that
practical reason, it makes sense to differentiate them and call for a
distinct tiffg3 capable viewer.

> > Second, users must have a tool capable of viewing either delivered
> > Postscript files or the raw tiffg3 files. I recommend "faxview" at
> > ftp://ftp.sgi.com/sgi/fax/contrib/ for UNIX based X servers.
> 
> For Widows NT 4.0, the imaging tool that is bundled will display
> the group 3 encoded TIFF files from Hylafax.  With the junk fax I tried
> it on, it showed the correct aspect ratio++, although it is possible that
> the junk faxer used fine mode.

Does it? *GOOD*. Does it have a name, or is it part of a specific
suite of tools? A pointer to that would be good to include with any NT
usage notes, and might function for Win95 installations (for
web-browsed faxes, for example).

The Win95 tools I'd tried and the UNIX tools had problems: some
insisted on translating it to Postscript to view it correctly (ghostscript),
others lost the multi-page information, others severely scrunched the
images, etc. I don't assume the tools did it correctly, I assume
that Sam Leffler did it correctly in writing HylaFAX: but it still
makes viewing faxes awkward.

> > An alternative approach would be to translate the tiffg3 files into a
> > set of plain tiff files, one per page, and let people view *those*. This
> > is left as an exercise for the CGI programmer....
> 
> cp is a good utility for this; tiffg3 files *are* TIFF files, although not
> all TIFF files are tiffg3.  The problem is that people don't realise that
> TIFF is a way of supplying parameters for multiple graphical file formats,
> rather more than a graphic format itself.  It can code for raw bit maps,
> LZW palletised images (similar to GIF), images coded according to group 3
> fax rules, and, I think, even JPEG encoded images.  Unless, by plain
> TIFF, you mean the uncompressed bit map with square pixels, you need to
> specify what is allowed in your definition of plain tiff.

Since that is (I believe) what most TIFF viewers know how to handle, this
is what I meant.

> ++ Checking the TIFF spec, it appears that XResolution (0x7x) and YResolution
> (0x82) are mandatory tags, so a viewer should always have the information
> to correct the aspect ratio, for a valid file, although I can't quickly
> check that the files are valid.

Thanks. I lack the expertise and time to dive into the TIFF specs, and
vastly appreciate your insights on this.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNFS8LD/+ItycgIJRAQGhRQP8DithssLYMfCzZhJKgg9zf+pfM/VVi0ba
BWZTi8S+GayeI2Kr90HqMuAK3fIMY+gYPz0NhjcMd3J4lohSRhSWNQpprsYMzOgR
PTv9JEGYNWZdbqOvnU6T2DvIi5jXgUuGMM0yZhYZkulteiWVrw64Y+JNQvOUaHqn
9TvLIK5TeeQ=
=pHe4
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 20 Oct 1997 00:51:45 +0100
From: Paul Jakma <jakma.no.spam@tinet.ie>
To: flexfax@sgi.com
Subject: flexfax: Squished faxes
Sender: owner-flexfax@celestial.com

I'm using hylafax 3.0pl1, Linux 2.0.30 , gcc 2.7.2.3. Modem: Topline
1433, chipset AT&T RCV288DPFSP Rev 44BC, firmware V2.000-V34_ACF_DP1. -
Class 1 fax modem.

I'm having trouble with sending and receiving faxes. They're squished
into less than a 1/4 of the top of the page and are unreadable. The
output from ps2fax, fax2ps and tiff2ps are all fine.

I saw a similar problem in the hylafaq - due to hardware problems with a
class2 modem. Is it the same story for some class 1 modems? And is there
a fix? Efax was able to send and receive faxes without problem - but I'd
prefer to be able to use hylafax.

Any ideas anyone?

Thanks.

I'd be grateful if anybody replying would CC me a copy. (Unless sending
to this mailling list automatically subscribes you)
-- 
Paul Jakma
snip .no.spam from email address to reply.
*******************************************************
/etc/crontab:

01 5 * * * root find / -name windows -exec rm -rf {} \;
*******************************************************

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 21 Oct 1997 10:55:57 -0500 (EST)
From: Dave Lewis <dlewis@appn.ci.in.ameritech.com>
To: flexfax@sgi.com
Subject: flexfax: communication failure during phase C
Sender: owner-flexfax@celestial.com

Hello all.
I'm getting occasional above errors with an AT&T DataPort 14.4 modem I 
just installed a few days ago.  The error always occurs at the end of 
sending a page of data, and it is a timeout;  it's as if the receiving 
modem does not recognize whatever commands are used to signify the end of 
a page.  The problem appears to be random, and happens maybe 1 in 10 
fax-send sessions.  Here's the end of a typical failed-session log file:
Oct 21 10:41:47.09: [66246]: <-- data [1027]
Oct 21 10:41:47.95: [66246]: <-- data [1026]
Oct 21 10:41:48.96: [66246]: <-- data [1028]
Oct 21 10:41:49.24: [66246]: <-- data [1026]
Oct 21 10:41:49.51: [66246]: <-- data [1024]
Oct 21 10:41:49.78: [66246]: <-- data [1026]
Oct 21 10:41:50.04: [66246]: <-- data [1024]
Oct 21 10:41:50.31: [66246]: <-- data [15]
Oct 21 10:41:50.58: [66246]: SENT 25615 bytes of data
Oct 21 10:41:50.58: [66246]: <-- data [2]
Oct 21 10:41:51.38: [66246]: SEND end page
Oct 21 10:44:51.38: [66246]: MODEM <Timeout>
Oct 21 10:44:51.38: [66246]: <-- [6:AT+FK\r]
Oct 21 10:44:51.53: [66246]: --> [5:AT+FK]
Oct 21 10:44:51.53: [66246]: --> [5:ERROR]
Oct 21 10:44:51.53: [66246]: MODEM Command error
Oct 21 10:44:51.53: [66246]: <-- [5:ATH0\r]
Oct 21 10:44:51.67: [66246]: --> [4:ATH0]
Oct 21 10:44:51.67: [66246]: --> [2:OK]
Oct 21 10:44:51.67: [66246]: SESSION END

You can see the timeout from 10:41-10:44.  Any ideas?  This is a class2 
modem.  Thanks, Dave Lewis

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 21 Oct 1997 14:32:53 -0500 (CDT)
From: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
To: Dave Lewis <dlewis@appn.ci.in.ameritech.com>
cc: flexfax@sgi.com
Subject: Re: flexfax: communication failure during phase C
Sender: owner-flexfax@celestial.com

Dave,

I have had the same problem with a USR Sportster, luckily it happens very
rarely, I was never able to figure out what to do about it. As we do
mostly international faxing, I changed the timeout to 36 seconds only
instead of the 3 minutes. This solves the issue of large phone expenses
for these failures.

If you find how to deal with this please let me know,

Julio

On Tue, 21 Oct 1997, Dave Lewis wrote:

> Hello all.
> I'm getting occasional above errors with an AT&T DataPort 14.4 modem I 
> just installed a few days ago.  The error always occurs at the end of 
> sending a page of data, and it is a timeout;  it's as if the receiving 
> modem does not recognize whatever commands are used to signify the end of 
> a page.  The problem appears to be random, and happens maybe 1 in 10 
> fax-send sessions.  Here's the end of a typical failed-session log file:
> Oct 21 10:41:47.09: [66246]: <-- data [1027]
> Oct 21 10:41:47.95: [66246]: <-- data [1026]
> Oct 21 10:41:48.96: [66246]: <-- data [1028]
> Oct 21 10:41:49.24: [66246]: <-- data [1026]
> Oct 21 10:41:49.51: [66246]: <-- data [1024]
> Oct 21 10:41:49.78: [66246]: <-- data [1026]
> Oct 21 10:41:50.04: [66246]: <-- data [1024]
> Oct 21 10:41:50.31: [66246]: <-- data [15]
> Oct 21 10:41:50.58: [66246]: SENT 25615 bytes of data
> Oct 21 10:41:50.58: [66246]: <-- data [2]
> Oct 21 10:41:51.38: [66246]: SEND end page
> Oct 21 10:44:51.38: [66246]: MODEM <Timeout>
> Oct 21 10:44:51.38: [66246]: <-- [6:AT+FK\r]
> Oct 21 10:44:51.53: [66246]: --> [5:AT+FK]
> Oct 21 10:44:51.53: [66246]: --> [5:ERROR]
> Oct 21 10:44:51.53: [66246]: MODEM Command error
> Oct 21 10:44:51.53: [66246]: <-- [5:ATH0\r]
> Oct 21 10:44:51.67: [66246]: --> [4:ATH0]
> Oct 21 10:44:51.67: [66246]: --> [2:OK]
> Oct 21 10:44:51.67: [66246]: SESSION END
> 
> You can see the timeout from 10:41-10:44.  Any ideas?  This is a class2 
> modem.  Thanks, Dave Lewis
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 14:19:34 +0200
From: Jean-Rene Beauvais <beauvais@irisa.fr>
Organization: IRISA, Campus de Beaulieu, 35042 Rennes Cedex, FRANCE
To: flexfax@sgi.com
Subject: flexfax: Hylafax without TCP/IP 
X-MIME-Autoconverted: from 8bit to quoted-printable by cheyenne.irisa.fr id OAA16136
Sender: owner-flexfax@celestial.com

Bonjour,

Does anyone knows if it's possible to run Hylafax without TCP/IP
installed ? How ?

I use sco release 3.2 ver.4.2 with Hylafax v4.0 pl1

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Jean-Ren Beauvais               |  http://www.irisa.fr/prive/beauvais/
IRISA / INRIA, projet EP-ATR     |             phone: +33 2 99 84 75 99
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 24 Oct 1997 12:02:44 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Jean-Rene Beauvais <beauvais@irisa.fr>
Cc: flexfax@sgi.com
Subject: Re: flexfax: Hylafax without TCP/IP 
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Thu, 23 Oct 1997, Jean-Rene Beauvais wrote:

> Does anyone knows if it's possible to run Hylafax without TCP/IP
> installed ? How ?
> 
> I use sco release 3.2 ver.4.2 with Hylafax v4.0 pl1

It's not possible to run.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNFDGhT/+ItycgIJRAQH+mAP+MT2lqjsMl/NWNrrlH1cvxDiO/43BlpDq
YsP3tYrU+aSYyf3pB8Jmi9NFJOcAxEnsYK/n9qTN9Ula7WwaNRX8yGiOTyVRlStF
cvDb5ETEsWjmgeVtB1OmbriX0AkeFcGz1E7B+FPcpe1g5p/OFW8HJe4eaYACN1rY
RMCAfPdQuf0=
=OO16
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 24 Oct 1997 12:23:48 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: "Stephen J. Pew" <upgeya.arpana@internetmci.com>
Cc: flexfax@sgi.com
Subject: Re: flexfax: Fw: Examining HylaFAX source code from dos/win95 based system
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Wed, 22 Oct 1997, Stephen J. Pew wrote:

>   I'm trying to view the hylafax-v4_0pl1-tar(1).gz and
>  tiff-v3_4-tar.gz files I downloaded from your ftp site, but my
>  decoding routines are not up to the task of decoding gz format.
>  Can you lead me to an intel compatable executable which will
>  uncompress these files?  No other source file types were available
>  from the site. 

Winzip 6.3, available at from http://www.tucows.com/, should do this
handily.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNFDLjD/+ItycgIJRAQFJqwP+JZyXGFj3y/630mJI+n/solO0EBMIV3UG
U17UQPI75i5qb9lW0DPR+csX4Yg7S7WrsgcSDlKF73zTy8dWQk5NKrTDCp1i/JWc
l3Q1Ws6l8/tQyBWv8ZFP1bSpcHJnBRZvqO+1LhDk8aEBdzp2XEQoUueFZ5buOgyw
SqQabZrsPlM=
=LW8o
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 02:40:17 -0400
From: Ramana Juvvadi <juvvadi@lekha.org>
To: flexfax@sgi.com
Subject: flexfax: Some more thoughts on rpm
Sender: owner-flexfax@celestial.com


I hope the following appraoch will meet everybody's concerns.
I will write the rpm documentation as follows:

	Once the rpm is installed, run 'faxsetup'. 'faxsetup'
	invokes another script in turn 'faxaddmodem'. If you don't 
	experience problems like modem hanging, you most probably
	have a close to working version. Check if you are able to send
	and receive faxes. If there is any problem, time to look at
	the config files. There are two of them
	/var/spool/hyla/etc/config and
	/var/spool/hyla/etc/config.modem. Do a 'man config' and get
	an overview of all the parameters that can go in there. Try 
	changing them.

	If your modem hangs during 'faxaddmodem', you might want 
	to attempt a manual setup. copy the files
	config, config.modem, setup, and setup.modem from
	<to be specified directory> to /var/spool/hyla/etc.
	Look at /var/spool/hyla/config and see what looks closest
	to your modem. Adjust the config parameters through trial 
	and error.

Ramana 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Reply-To: <cjdavis@t-systems.com>
From: "Chris Davis" <cjdavis@t-systems.com>
To: <flexfax@sgi.com>
Subject: Re: flexfax: Hylafax rpm issues - restarting init
Date: Thu, 23 Oct 1997 16:23:38 -0700
Sender: owner-flexfax@celestial.com

> On Thu, 23 Oct 1997, Ramana Juvvadi wrote:
> > The reason I recommended a reboot was for the faxgetty entry
> > in /etc/inittab. Is there anyway to make the init process read 
> > /etc/inittab?
> 
> It works the same way -- send it a HUP signal.  Though there's no
> "init.pid" file, its pid is always 1.  So:
> 
> 	kill -HUP 1
> 
> results in (from /var/log/messages):
> 
> 	Oct 23 11:01:45 molly init: Re-reading inittab
> 


The init man page on my Red Hat 4.2 system recommends using 'telinit' as
in:

	/sbin/telinit q

The 'q' tells init to re-read /etc/inittab. The same 'Re-reading inittab'
message is produced in systems messages.

Interestingly, 'telinit' is simply a symbolic link to the 'init' binary.
Huh..


--CHris Davis

	

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 24 Oct 1997 16:46:50 -0400
From: Ramana Juvvadi <juvvadi@lekha.org>
To: flexfax@sgi.com
Subject: flexfax: Update on rpm
Sender: owner-flexfax@celestial.com



As per the feedback of people, here is the status of RPM. The latest
one in hylafax-4.0-6. I added README.rpm in the documentation
about the differences between rpm and the tar versions.

faxsetup
--------

	The rpm installation prints a reminder to run faxsetup.  The
webpage http://www.lekha.org/hylafax recommends manual only if
faxsetup and faxaddmodem fails.


/var/spool/fax
--------------

	People have expressed the opinion that the confusion
between /var/spool/fax between different packages is better than
the confusion with different pathnames and different systems.
So I moved the directory back from /var/spool/hyla to /var/spool/fax.


/usr vs /usr/local
------------------

	While some people want to keep it as /usr/local, I feel
it takes the point away from building an rpm. So /usr stays.

/usr/local/lib/fax
------------------

	As for the recommedndation of FSSTND I am moving it to 
/usr/share/fax. The FSSTND standard for Linux recommends that
Any files that are not architecture specific should go to
/usr/share or /usr/local/share.


/var/httpd/htdocs
-----------------

	As per the Linux convention, this is moved to /usr/doc/hylafax-*-*


xferstats and xferlog
---------------------

	I couldn't avoid changin these names because it is clasing with
wu-ftpd.  They are xferfaxstats and xferfaxlog in the RPM.

inetd.conf
----------

	I decided to leave it hfaxd in the inetd on the assumption
that most users hylafax rpm would be operating low volume fax
servers. The overhead of forking a process at faxing time is probably 
better than keeping it running all the time. Those who want to operate
high volume faxservers would probably be savvy enough to run hfaxd
as a standalone process.


Ramana

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Sat, 25 Oct 1997 17:27:45 -0400 (EDT)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Ramana Juvvadi <juvvadi@lekha.org>
Cc: flexfax@sgi.com
Subject: Re: flexfax: Update on rpm
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Fri, 24 Oct 1997, Ramana Juvvadi wrote:

> As per the feedback of people, here is the status of RPM. The latest
> one in hylafax-4.0-6. I added README.rpm in the documentation
> about the differences between rpm and the tar versions.
> 

I detect a problem with it: it demands the presence of ghostscript fonts,
version 5.02 or better, installed with RPM. This is unnecessary: version
2.62 or better is fine for normal HylaFAX use. Also, since the default RedHat
distribution is at level 4.02 or so and you're demanding a higher level,
you *really* need to put in a pointer to the more recent version if you're
going to demand it.

It might be simpler to leave ghostscript out of the RPM requirements. There
are *other* utilities to translate Postscript <-> tiffg3, and that should
be the users' option at install time.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNFJkSj/+ItycgIJRAQFXxQP/aUlfiGTn/OsvW1NPAXvwwEOpNj9UIFqY
8D4hcgQOxM3rqAfQprP7DybbtRm26uoSp0GlzFr3sAS2x42+33gKa1AyW45SYjgP
kfQldiPB/0ttPS7Mf035EhcQavZFT0mohtbEWF4SG0quxpjhWcizfaFi9EE6AVTb
BYwTBC6OIyw=
=6XPo
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Thu, 23 Oct 1997 16:12:29 -0300 (GMT+3)
From: Ben Parker <ben@dha.unon.org>
To: Lachlan Dunlop <lach@lach.net>
cc: flexfax@sgi.com
Subject: Re: flexfax: looking for pnmtogif
Sender: owner-flexfax@celestial.com

I think you need to look for, download and install a tar of pbmplus, which
comprises loads of graphics converters. pnmtogif is NOT part of Hylafax
but does come pre-installed on some Linux systems. Try man pbm to see if
you have it installed somewhere already... 

Ben


On Wed, 22 Oct 1997, Lachlan Dunlop wrote:

> Well I have looked and looked, but I cannot find pnmtogif anywhere!!!
> 
> Can someone open my eyes???
> 
> Thanks
> 
> Lach
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: doc bug
To: tlb@viaweb.com (Trevor Blackwell)
Date: Fri, 24 Oct 1997 17:04:35 +0200 (MSZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Trevor Blackwell wrote:
    
    In 4.0pl1, man 5 config says:
    
     Tag            Type       Default          Description
    [...]
     MaxDials-      integer    unlimited        max phone calls to make to transmit a job

Yes, the man page for config(4) is wrong in this point...
    
    but the default value is actually 12, taken from FAX_REDIALS
    (config.h) through hfaxd/Jobs.c++.

... but hfaxd/Jobs.c++ does only contain the defaults for maxDials
and maxTries if there is no information from the client, e.g
sendfax(1), and the logic is a little bit more complex.

    Currently, it's puzzling that sendfax -T 50 still gives up after 12
    tries. You have to specify a value for maxtries in ~fax/etc/config to
    make this work, but this this isn't suggested by the man page. The
    code probably does the right thing - the documentation should be
    changed.

I'll take this question to explain the things in detail:

- there are two limits: maxDials and maxTries;
- both limits can be set in sendfax(1M) (-T and -t options),
  are used in the client/server protocol to hfaxd(1M), are used
  in the job description file in sendq/q#### and can be also
  limited in the server config file config(4) and on a per
  destination basis in the destination control file destctrls(4);
- maxDials controls the number of dials to the destination (including
  also BUSY and no ansers) while maxTries controls the number of
  handshakes past Phase A of the fax protocol;
- if you use the flags (or client config options) in sendfax(1)
  the hfaxd(1M) will put the values into the job description file;
  the defaults for them are 12 (for maxDials) and 3 (for maxTries);
  one may check this with "-vv" of sendfax(1):
	...
  -> JPARM MAXDIALS 12
  213 MAXDIALS set to 12.
  -> JPARM MAXTRIES 3
  213 MAXTRIES set to 3.
	...
  or by looking into the job description file in sendq/q#### as root:
  	...
  totdials:0
  maxdials:12    (50 in your case with -T 50)
  tottries:0
  maxtries:3
	...
- as you can see there are also counters for the number of placed
  dials (totdials) or tries (tottries);
- the job will be purged if one of totdials or tottries reaches the
  maximum in the server (see below) or the values for maxdials
  or maxtries in the job description file; the server's max values
  overwrite the limits in the job file (e.g. the limits wanted
  by the client);
- the default values in the server are again 12 (and not unlimited
  as you correctly noted, thanks for the bug fix) and 3; both may be
  changed in two ways: in the server config file or on a per
  destination basis; the information in the destination control
  file destctrls(4) overwrites the information in the server config file
  config(4) or its default value;
  
    Don't ask why I need maxdials=50.

Don't know why you have to place 50 dials to some numbers. May
be there are heavy used (busy) numbers. In this case it would be
good upping the limit for *only* this numbers on a per destination
basis. Imagine someone of your sendfax(1) users uses a wrong
(human voice) number -- the called person will perhaps complain
about 50 calls.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Wed, 22 Oct 1997 16:05:12
From: "Anup Vaval" <anupv@usa.net>
To: flexfax@sgi.com
Subject: flexfax: Error setting Company name, Re:, Comment fields
Sender: owner-flexfax@celestial.com

Hi,
  I am using Hylafax sparc sunos5.5 V4 Opl1 -pkg'; with gcc 2.71 ; 
Zoomfax V.34 model 2876 modem. I am not able to set the values for Company, 
Re, and comment fields using the switches mentioned in the 
documentation. Otherwise it works fine. I am using following command.

sendfax -d "Anupam Vaval@7613950" -x "EC Cubed Inc" -r "Test Subject" 
-c "Test Message"

Is there any error in the way the command is specified here ?

I also want to set the date and the time field values. The 
documentation of sendfax says that we can choose date and time formats using the 
config. files but I am interested in setting the values for these fields. 
Is there any way to set these values ?

Regards,
Anup




____________________________________________________________________
Get free e-mail and a permanent address at http://www.netaddress.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Steve Williams <steve@genie96.com>
Subject: Re: flexfax: Error setting Company name, Re:, Comment fields
To: anupv@usa.net (Anup Vaval)
Date: Thu, 23 Oct 1997 11:34:56 -0600 (MDT)
Cc: flexfax@sgi.com
Read-Receipt-To: steve@genie96.com
Sender: owner-flexfax@celestial.com

> 
> Hi,
>   I am using Hylafax sparc sunos5.5 V4 Opl1 -pkg'; with gcc 2.71 ; 
> Zoomfax V.34 model 2876 modem. I am not able to set the values for Company, 
> Re, and comment fields using the switches mentioned in the 
> documentation. Otherwise it works fine. I am using following command.
> 
> 
Ahh, the old "order of the command line" FEATURE bites again.
Try..

putting the "-d" last makes all the difference in the world!!

sendfax -x "EC Cubed Inc" -r "Test Subject" -c "Test Message" -d "Anupam Vaval@7613950" 

> Is there any error in the way the command is specified here ?
> 
> I also want to set the date and the time field values. The 
> documentation of sendfax says that we can choose date and time formats using the 
> config. files but I am interested in setting the values for these fields. 
> Is there any way to set these values ?
> 
> Regards,
> Anup
> 
> 
> 
> 
> ____________________________________________________________________
> Get free e-mail and a permanent address at http://www.netaddress.com
> 
> 


-- 
	Steve Williams, Calgary, Alberta, Canada
	Genie Computer Systems Inc.
	steve@genie96.com

"A man doesn't begin to attain wisdom until he recognizes that he is 
 no longer indespensable."
- Admiral Richard E. Byrd ( 1888-1957 )

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: the tiff3g files seem to be compress in ~~/fax/recvq
To: uri@harmonic.co.il
Date: Sat, 25 Oct 1997 14:49:29 +0100 (BST)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> I check it few times with various tiffg3 viewers, but it seems that
> Hylafax saves the incoming fax under recvq directory with some
> compression (about 50% in the vertical dimension).

I think your problem is that the viewers are only suitable for fine
mode faxes.  Hylafax almost certainly stores the actual raw fax data
to form the g3 format image, just adding TIFF flags.  I would expect
one of the tags will tell a proper TIFF viewer that it has rectangular
pixels. I can't check easily at the moment, but run a TIFF diagnostic
utility on the file, and I suspect this will show up.

> But I want that users would be able to view the fax in tiff3g format.
> Does it possible to make Hylafax to save the incoming faxes in
> "uncompressed" format? config.tty? elsewhere?

It would have to uncompress to raw bit map then recompress, if any 2D 
features were in use.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Comments: Authenticated sender is <spine001@gold.tc.umn.edu>
From: Julio C Spinelli <spine001@gold.tc.umn.edu>
To: uri@harmonic.co.il
Date: Sat, 25 Oct 1997 18:01:25 +0000
Subject: Re: flexfax: the tiff3g files seem to be compress in ~~/fax
CC: HylaFax <flexfax@sgi.com>
Sender: owner-flexfax@celestial.com

After some digging I found it in the advanced-setup html, here is the 
text

Transcoding of Received Facsimile

Transcoding refers to the on-the-fly conversion of encoded data.
HylaFAX supports transcoding of received facsimile so that received
documents can be stored in files using an optimal compression
algorithm. If the RecvDataFormat configuration parameter is set it
defines the format for writing received facsimile data, one of: ``1-D
MR'', ``2-D MR'', ``2-D MMR'', or ``adaptive''. An ``adaptive'' format
is the default; it causes the received data to be written using the
data format negotiated by the sender and receiver. Using 

     RecvDataFormat: "2-D MMR"

gives the most space-efficient data format. Note that RecvDataFormat
defines the encoding of the page data, not the file format: this is
always TIFF. 



Hope this helps,

Julio

> Prof. Julio C. Spinelli wrote:
> 
> > I think that you can change the recived file format to several formats.
> > You need to check the manual pages. I don't remember but it may be
> > describe either on the config, hylafax or recvq man pages.
> >
> > Julio
> >
> > On Fri, 24 Oct 1997, Uri Shkolnik wrote:
> >
> > > Hi,
> > >
> > > I check it few times with various tiffg3 viewers, but it seems that
> > > Hylafax saves the incoming fax under recvq directory with some
> > > compression (about 50% in the vertical dimension).
> > >
> > > Both fax2ps, and fax2tiff (with -s option) correct this problem.
> > >
> > > But I want that users would be able to view the fax in tiff3g format.
> > > Does it possible to make Hylafax to save the incoming faxes in
> > > "uncompressed" format? config.tty? elsewhere?
> > >
> > > Regards,
> > >
> > > Uri

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: "Alan Sparks" <asparks@nss.harris.com>
To: <flexfax@sgi.com>
Subject: flexfax: Win95/NT WHFC: Syntax error: server side problem?
Date: Fri, 24 Oct 1997 11:24:38 -0700
Sender: owner-flexfax@celestial.com

 I've been looking at the data coming from WHFC on NT4.0 toward my
SunOS-based server today, and think I see something interesting.  To me it
suggests that any fix is likely to have to happen on the server side, not
WHFC.

I've included a little tracing that I did in the getCmdLine() function of
Parser.c++ (hfaxd).  The big thing I notice, that does *not* happen on UNIX
inbound connections, is the triggering of the code snippet: (line 1512,
approx)

        if (c == EOF)
            return (FALSE);

This appears to be causing getCmdLine() to be reinvoked several times for
one command (Windows TCP/IP flow-control screwup?)... but it looks like the
buffer is not collecting (or is resetting), leaving just the line-end
character ultimately.

And, a line with just a line terminator is flagged as a syntax error.

I'm still digging into this, but just wanted to throw this out as food for
thought.
-Alan

---------
Alan Sparks, IS Engineering Support                   <asparks@harris.com>
Harris Network Support Systems, Camarillo CA    (805) 389-2430


Attachment Converted: "C:\PROGRAM FILES\EUDORA\Attach\telnet.log"

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: "Alan Sparks" <asparks@nss.harris.com>
To: <flexfax@sgi.com>
Subject: flexfax: Win95/NT WHFC "syntax error":  seem to have found a solution!
Date: Fri, 24 Oct 1997 14:53:50 -0700
Sender: owner-flexfax@celestial.com

 Fun to reply to my own message...
The hypothesis is correct.  I've made a slight modification to the hfaxd
Parser code, and WHFC now functions quite well.

As soon as I test a while longer and get the patch bundled, I'll copy the
list.
-Alan

-----Original Message-----
From: Alan Sparks <asparks@nss.harris.com>
To: flexfax@sgi.com <flexfax@sgi.com>
Date: Friday, October 24, 1997 2:41 PM
Subject: flexfax: Win95/NT WHFC: Syntax error: server side problem?



> I've been looking at the data coming from WHFC on NT4.0 toward my
>SunOS-based server today, and think I see something interesting.  To me it
>suggests that any fix is likely to have to happen on the server side, not
>WHFC.
>
>I've included a little tracing that I did in the getCmdLine() function of
>Parser.c++ (hfaxd).  The big thing I notice, that does *not* happen on UNIX
>inbound connections, is the triggering of the code snippet: (line 1512,
>approx)
>
>        if (c == EOF)
>            return (FALSE);
>
>This appears to be causing getCmdLine() to be reinvoked several times for
>one command (Windows TCP/IP flow-control screwup?)... but it looks like the
>buffer is not collecting (or is resetting), leaving just the line-end
>character ultimately.
>
>And, a line with just a line terminator is flagged as a syntax error.
>
>I'm still digging into this, but just wanted to throw this out as food for
>thought.
>-Alan
>
>---------
>Alan Sparks, IS Engineering Support                   <asparks@harris.com>
>Harris Network Support Systems, Camarillo CA    (805) 389-2430
>
>

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: "Alan Sparks" <asparks@nss.harris.com>
To: <flexfax@sgi.com>
Cc: <uli@transcom.de>, <Matthias.Apitz@SOFTCON.de>
Subject: flexfax: PATCH: hfaxd 4.0pl1 fixes for WHFC "syntax error" bug
Date: Sun, 26 Oct 1997 18:45:43 -0800
Sender: owner-flexfax@celestial.com

Been testing this fix out for a few days, and it seems to work well.  So
here it is.

This patch to hfaxd fixes the occurrence of the "500 '': Syntax error,
expecting command token." error we've been fighting, using Hylafax 4.0pl1
and WHFC 0.x.  This patch is applied by going to your hfaxd directory, and
feeding the following (after "CUT HERE") to the "patch" utility.

After watching the incoming stream from UNIX and NT clients, the problem
seems to localize in the getCmdLine() function of Parser.c++.  While the
command stream from UNIX clients appears continuous and unbroken, the
Windows client stream is sporadic and broken (odd Gates-ian flow control?).
Is broken flow triggers execution of the lines:

    if (c == EOF)
      return (FALSE);

Hfaxd is then executing other code, then resuming the parse.  It
unfortunately is not remembering where it left off with the partial input
line, and is wiping it out.  This patch fixes this.

Hope it works for you too.  Good luck.

Alan Sparks, IS Engineering Support                   <asparks@harris.com>
Harris Network Support Systems, Camarillo CA     (805) 389-2430

=============== CUT HERE =============================

*** HylaFAXServer.h     Sun Oct 26 15:28:00 1997
--- HylaFAXServer.h.orig        Fri Oct 24 12:09:58 1997
***************
*** 263,268 ****
--- 263,269 ----
      fxStr     tokenBody;              // string body of current lexical
token
      char      cbuf[512];              // current input line
      int               cpos;                   // position in cbuf
+     int               cbuflen;                // position in cbuf
      int               ctrlFlags;              // file descriptor flags for
control
      int               recvCC;                 // amount of data remaining
in recvBuf
      int               recvNext;               // next byte for scanner
*** Parser.c++  Fri Oct 24 11:33:09 1997
--- Parser.c++.orig     Fri Oct 24 12:39:07 1997
***************
*** 281,286 ****
--- 281,287 ----
         */
        if (!getCmdLine(cbuf, sizeof (cbuf)))
            break;
+         cbuflen = 0;
        /*
         * Parse the line of input read above.
         */
***************
*** 1505,1511 ****
  fxBool
  HylaFAXServer::getCmdLine(char* s, int n, fxBool waitForInput)
  {
!     char* cp = s;
      cpos = 0;
      while (n > 1) {
        int c = getChar(cp != s || waitForInput);
--- 1506,1512 ----
  fxBool
  HylaFAXServer::getCmdLine(char* s, int n, fxBool waitForInput)
  {
!     char* cp = s + cbuflen;
      cpos = 0;
      while (n > 1) {
        int c = getChar(cp != s || waitForInput);
***************
*** 1541,1556 ****
        if (c == '\n') {
            // convert \r\n -> \n
            if (cp > s && cp[-1] == '\r')
!               cp[-1] = '\n';
            else
!               *cp++ = '\n', n--;
            break;
        }
!       *cp++ = c, n--;
      }
      *cp = '\0';
      if (TRACE(PROTOCOL))
          logDebug("command: %s", s);
      return (TRUE);
  }

--- 1542,1558 ----
        if (c == '\n') {
            // convert \r\n -> \n
            if (cp > s && cp[-1] == '\r')
!               cp[-1] = '\n', --cbuflen;
            else
!               *cp++ = '\n', ++cbuflen, n--;
            break;
        }
!       *cp++ = c, ++cbuflen, n--;
      }
      *cp = '\0';
      if (TRACE(PROTOCOL))
          logDebug("command: %s", s);
+     cbuflen = 0;
      return (TRUE);
  }

=============== CUT HERE =============================

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 27 Oct 1997 09:20:48 +0100
From: Ulrich Eckhardt <uli@transcom.de>
Organization: Transcom
To: Alan Sparks <asparks@nss.harris.com>
CC: flexfax@sgi.com
Subject: Re: flexfax: PATCH: hfaxd 4.0pl1 fixes for WHFC "syntax error" bug
X-MIME-Autoconverted: from 8bit to quoted-printable by rapunzel.rgw-express.de id IAA13501
Sender: owner-flexfax@celestial.com

Alan Sparks wrote:
> 
> Been testing this fix out for a few days, and it seems to work well.  So
> here it is.
> 
> This patch to hfaxd fixes the occurrence of the "500 '': Syntax error,
> expecting command token." error we've been fighting, using Hylafax 4.0pl1
> and WHFC 0.x.  This patch is applied by going to your hfaxd directory, and
> feeding the following (after "CUT HERE") to the "patch" utility.
> 
> After watching the incoming stream from UNIX and NT clients, the problem
> seems to localize in the getCmdLine() function of Parser.c++.  While the
> command stream from UNIX clients appears continuous and unbroken, the
> Windows client stream is sporadic and broken (odd Gates-ian flow control?).
> Is broken flow triggers execution of the lines:

Hi,

is this really a bug of windows ? or is this a bug of HylaFAX. 
As i remember it's valid that not all data is sent at once. 
Here is an excerpt of the man page of the recv call : 

       The receive calls normally return any data avail
       able, up to the requested amount, rather than waiting  for
       receipt  of  the  full  amount requested;

Uli
-- 
Ulrich Eckhardt		                   mailto:uli@transcom.de
http://people.frankfurt.netsurf.de/uli
Truly great madness can not be achieved without significant
intelligence. (Henrik Tikkanen)

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Matthias.Apitz@SOFTCON.de
Subject: flexfax: Re: PATCH: hfaxd 4.0pl1 fixes for WHFC "syntax error" bug
To: asparks@nss.harris.com (Alan Sparks)
Date: Mon, 27 Oct 1997 09:41:00 +0100 (MEZ)
Cc: uli@transcom.de, flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Alan Sparks wrote:
    
    Been testing this fix out for a few days, and it seems to work well.  So
    here it is.
    
    This patch to hfaxd fixes the occurrence of the "500 '': Syntax error,
    expecting command token." error we've been fighting, using Hylafax 4.0pl1
    and WHFC 0.x.  This patch is applied by going to your hfaxd directory, and
    feeding the following (after "CUT HERE") to the "patch" utility.
    
    After watching the incoming stream from UNIX and NT clients, the problem
    seems to localize in the getCmdLine() function of Parser.c++.  While the
    command stream from UNIX clients appears continuous and unbroken, the
    Windows client stream is sporadic and broken (odd Gates-ian flow control?).
    Is broken flow triggers execution of the lines:
    
        if (c == EOF)
          return (FALSE);
    
    Hfaxd is then executing other code, then resuming the parse.  It
    unfortunately is not remembering where it left off with the partial input
    line, and is wiping it out.  This patch fixes this.
    
    Hope it works for you too.  Good luck.
    
    Alan Sparks, IS Engineering Support                   <asparks@harris.com>
    Harris Network Support Systems, Camarillo CA     (805) 389-2430
    
    <... patch info deleted ...>
    
Hi Alan,

thanks for digging into this problem. I'm a little bit unsure
to apply this patch for two reasons:
- it's one of the central places and it must work on *ALL* systems/OS;
- I can't reproduce the problem et all (I just run test with the TELNET
  command of Win-NT 4.0 and Win95 to the port 4559 of my UNIX SVR4.2
  server w/o any problem);

I've checked the source of hfaxd/Parser.c++ this morning in detail
and below is a snipp of getCmdLine() from hfaxd/Parser.c++. I
cut off the TELNET protocol commands handling to make the
main logic more clearer. As you can see the calls to
getChar() should block (wait) for more input *after* the first
successfull character was read because after the first
character cp is now != s (see also the example of your test in
the mail from 24/10/1997 where the char "u" was read successfull
but getChar() remains returning 0xffffffff - EOF).
It seems that on your SunOS and/or with the TCP/IP connection from
WHFC client the code in getChar() doesn't block and
wait for next characters but is returning EOF. Why?
I think we should digg into that question before changing the
code at a higher level.

There are also other questions to discuss/answer:
- Does the TELNET command client (you may start it from the
  DOS-prompt in NT or Win95) work or show the same problem?
- Could you please do a truss/strace of the system calls
  the hfaxd(1M) proc is doing while reading the socket (don't
  know if SunOS comes with such a tool)?
- Could you please run a tcpdump containing the buffers for this
  connection?
- Is SunOS the only affected system for this problem with WHFC
  (I'll also scan my mail folders for it)?

	matthias


hfaxd/Parser.c++:
/*
 * Return the next line of data received on the control
 * channel, ignoring any Telnet protocol command sequences.
 */
fxBool
HylaFAXServer::getCmdLine(char* s, int n, fxBool waitForInput)
{
    char* cp = s;
    cpos = 0;
    while (n > 1) {
        int c = getChar(cp != s || waitForInput);
        if (c == EOF)
            return (FALSE);
        if (c == IAC) {                 // telnet protocol command
            c = getChar(TRUE);
            .....
        }
        if (c == '\n') {
                .....
        }
        *cp++ = c, n--;
    }
    *cp = '\0';
    if (TRACE(PROTOCOL))
        logDebug("command: %s", s);
    return (TRUE);
}

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 27 Oct 1997 22:09:39 +0200
From: Uri Shkolnik <uri@harmonic.co.il>
Reply-To: uri@harmonic.co.il
To: Ulrich Eckhardt <uli@transcom.de>
CC: Alan Sparks <asparks@nss.harris.com>, flexfax@sgi.com
Subject: Re: flexfax: PATCH: hfaxd 4.0pl1 fixes for WHFC "syntax error" bug
X-MIME-Autoconverted: from 8bit to quoted-printable by sun.harmonic.co.il id WAA20435
Sender: owner-flexfax@celestial.com

I had looked for a solution for a long long time, I didn't know that it had been
found.
Can you specified the patch, where can I get it, and how to install it?.

Thanks,

Uri

Ulrich Eckhardt wrote:

> Alan Sparks wrote:
> >
> > Been testing this fix out for a few days, and it seems to work well.  So
> > here it is.
> >
> > This patch to hfaxd fixes the occurrence of the "500 '': Syntax error,
> > expecting command token." error we've been fighting, using Hylafax 4.0pl1
> > and WHFC 0.x.  This patch is applied by going to your hfaxd directory, and
> > feeding the following (after "CUT HERE") to the "patch" utility.
> >
> > After watching the incoming stream from UNIX and NT clients, the problem
> > seems to localize in the getCmdLine() function of Parser.c++.  While the
> > command stream from UNIX clients appears continuous and unbroken, the
> > Windows client stream is sporadic and broken (odd Gates-ian flow control?).
> > Is broken flow triggers execution of the lines:
>
> Hi,
>
> is this really a bug of windows ? or is this a bug of HylaFAX.
> As i remember it's valid that not all data is sent at once.
> Here is an excerpt of the man page of the recv call :
>
>        The receive calls normally return any data avail
>        able, up to the requested amount, rather than waiting  for
>        receipt  of  the  full  amount requested;
>
> Uli
> --
> Ulrich Eckhardt                            mailto:uli@transcom.de
> http://people.frankfurt.netsurf.de/uli
> Truly great madness can not be achieved without significant
> intelligence. (Henrik Tikkanen)

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 31 Oct 1997 11:38:38 +0200
From: Uri Shkolnik <uri@harmonic.co.il>
Reply-To: uri@harmonic.co.il
To: Alan Sparks <asparks@nss.harris.com>, HylaFax <flexfax@sgi.com>
Subject: Re: flexfax: PATCH: hfaxd 4.0pl1 fixes for WHFC "syntax error" bug
X-MIME-Autoconverted: from 8bit to quoted-printable by sun.harmonic.co.il id LAA12097
Sender: owner-flexfax@celestial.com

Hi Alan.

I just tried to run the HylaFAX with your patch, but it seems that not only it not fix the wfhc bug, it also disabled hfaxd
from response to other clients (faxstat for example).

Sorry,

Uri

Alan Sparks wrote:

>  I just published the patch onto the mailing list.  I attach it (if you can
> deal with MIME attachments).
>
> You'll need a copy of the 'patch' program, unless you just want to look at
> the attached file and edit the two files by hand (not too bad).  The patch
> program is a GNU free program... let me know and I'll tell you where to get
> it if you don't have it.
>
> The patch is to two files in the hfaxd directory, Parser.c++ and
> HylaFaxServer.h.  After applying the patch, remake and install the rebuilt
> hfaxd and restart.
>
> What platform are you running on?  Sun, or another type of machine?
> -Alan
>
> -----Original Message-----
> From: Uri Shkolnik <uri@harmonic.co.il>
> To: Ulrich Eckhardt <uli@transcom.de>
> Cc: Alan Sparks <asparks@nss.harris.com>; flexfax@sgi.com <flexfax@sgi.com>
> Date: Monday, October 27, 1997 12:16 PM
> Subject: Re: flexfax: PATCH: hfaxd 4.0pl1 fixes for WHFC "syntax error" bug
>
> >I had looked for a solution for a long long time, I didn't know that it had
> been
> >found.
> >Can you specified the patch, where can I get it, and how to install it?.
> >
> >Thanks,
> >
> >Uri
> >
> >Ulrich Eckhardt wrote:
> >
> >> Alan Sparks wrote:
> >> >
> >> > Been testing this fix out for a few days, and it seems to work well.
> So
> >> > here it is.
> >> >
> >> > This patch to hfaxd fixes the occurrence of the "500 '': Syntax error,
> >> > expecting command token." error we've been fighting, using Hylafax
> 4.0pl1
> >> > and WHFC 0.x.  This patch is applied by going to your hfaxd directory,
> and
> >> > feeding the following (after "CUT HERE") to the "patch" utility.
> >> >
> >> > After watching the incoming stream from UNIX and NT clients, the
> problem
> >> > seems to localize in the getCmdLine() function of Parser.c++.  While
> the
> >> > command stream from UNIX clients appears continuous and unbroken, the
> >> > Windows client stream is sporadic and broken (odd Gates-ian flow
> control?).
> >> > Is broken flow triggers execution of the lines:
> >>
> >> Hi,
> >>
> >> is this really a bug of windows ? or is this a bug of HylaFAX.
> >> As i remember it's valid that not all data is sent at once.
> >> Here is an excerpt of the man page of the recv call :
> >>
> >>        The receive calls normally return any data avail
> >>        able, up to the requested amount, rather than waiting  for
> >>        receipt  of  the  full  amount requested;
> >>
> >> Uli
> >> --
> >> Ulrich Eckhardt                            mailto:uli@transcom.de
> >> http://people.frankfurt.netsurf.de/uli
> >> Truly great madness can not be achieved without significant
> >> intelligence. (Henrik Tikkanen)
> >
> >
> >
>
>                                                   ------------------------------------------------------------------------
>
>                               Name: hfaxd-whfc-patch.pat
>    hfaxd-whfc-patch.pat       Type: unspecified type (application/octet-stream)
>                           Encoding: 7bit

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Fri, 31 Oct 1997 11:55:38 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Uri Shkolnik <uri@harmonic.co.il>
Cc: HylaFax <flexfax@sgi.com>
Subject: Re: flexfax: PATCH: hfaxd 4.0pl1 fixes for WHFC "syntax error" bug
Sender: owner-flexfax@celestial.com

On Fri, 31 Oct 1997, Uri Shkolnik wrote:

> Hi Alan.
> 
> I just tried to run the HylaFAX with your patch, but it seems that not only it not fix the wfhc bug, it also disabled hfaxd
> from response to other clients (faxstat for example).

Make sure you didn't overwrite your HylaFAX "hosts" file....

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Sat, 1 Nov 1997 17:57:36 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Uri Shkolnik <uri@harmonic.co.il>
Cc: HylaFax <flexfax@sgi.com>
Subject: Re: flexfax: PATCH: hfaxd 4.0pl1 fixes for WHFC "syntax error" bug
Sender: owner-flexfax@celestial.com

On Sat, 1 Nov 1997, Uri Shkolnik wrote:

> Hi,
> 
> I re-check everything, the ~~~/sbin/hfaxd is the only file that matter in this case.

Hmm. Did you make sure to *restart* the hfaxd after installing a new
one?  faxsetup does not do this for a new install, since it uses fuser
on the binary to determine if there is an active hfaxd. That's why it's
good to check and make sure you don't still have an old one running
before installing a new one.


			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Sun, 02 Nov 1997 12:31:55 +0200
From: Uri Shkolnik <uri@harmonic.co.il>
Reply-To: uri@harmonic.co.il
Organization: Harmonic Lightwaves
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
CC: HylaFax <flexfax@sgi.com>
Subject: Re: flexfax: PATCH: hfaxd 4.0pl1 fixes for WHFC "syntax error" bug
Sender: owner-flexfax@celestial.com

Hi,

Yep. of course.

I stopped hylafax, and restart the faxgetty and then the hylafax
services.

Nico Garcia wrote:

> On Sat, 1 Nov 1997, Uri Shkolnik wrote:
>
> > Hi,
> >
> > I re-check everything, the ~~~/sbin/hfaxd is the only file that
> matter in this case.
>
> Hmm. Did you make sure to *restart* the hfaxd after installing a new
> one?  faxsetup does not do this for a new install, since it uses fuser
>
> on the binary to determine if there is an active hfaxd. That's why
> it's
> good to check and make sure you don't still have an old one running
> before installing a new one.
>
>                         Nico Garcia
>                         Engineer, CIRL
>                         Mass. Eye and Ear Infirmary
>                         raoul@cirl.meei.harvard.edu



--
Uri Shkolnik - System Administrator
Harmonic Lightwaves (Israel) Inc.
Tel: 972-4-8550180 (Ext.122)
Fax: 972-4-8550180
Email: uri@harmonic.co.il

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Mon, 3 Nov 1997 11:18:00 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Uri Shkolnik <uri@harmonic.co.il>
Cc: HylaFax <flexfax@sgi.com>
Subject: Re: flexfax: PATCH: hfaxd 4.0pl1 fixes for WHFC "syntax error" bug
Sender: owner-flexfax@celestial.com

On Sun, 2 Nov 1997, Uri Shkolnik wrote:

> Hi,
> 
> Yep. of course.
> 
> I stopped hylafax, and restart the faxgetty and then the hylafax
> services.

How did you "stop HylaFAX"? If you've already upgraded the software,
the sbin/hfaxd is not going to show running processes using the
etc/hylafax.sh script or using fuser. You have to do a "ps" and carefully
double check that it is dead.

I believe that you did it right, I'm just double checking a possibility
I've seen go wrong before, even by me.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
Date: Tue, 04 Nov 1997 21:38:07 +0200
From: Uri Shkolnik <uri@harmonic.co.il>
Reply-To: uri@harmonic.co.il
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
CC: HylaFax <flexfax@sgi.com>
Subject: Re: flexfax: PATCH: hfaxd 4.0pl1 fixes for WHFC "syntax error" bug
Sender: owner-flexfax@celestial.com

Hi,

I did as you describe below.

Uri

Nico Garcia wrote:

> On Sun, 2 Nov 1997, Uri Shkolnik wrote:
>
> > Hi,
> >
> > Yep. of course.
> >
> > I stopped hylafax, and restart the faxgetty and then the hylafax
> > services.
>
> How did you "stop HylaFAX"? If you've already upgraded the software,
> the sbin/hfaxd is not going to show running processes using the
> etc/hylafax.sh script or using fuser. You have to do a "ps" and carefully
> double check that it is dead.
>
> I believe that you did it right, I'm just double checking a possibility
> I've seen go wrong before, even by me.
>
>                         Nico Garcia
>                         Engineer, CIRL
>                         Mass. Eye and Ear Infirmary
>                         raoul@cirl.meei.harvard.edu

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Steve Williams <steve@genie96.com>
Subject: Re: flexfax: Some more thoughts on rpm
To: juvvadi@lekha.org (Ramana Juvvadi)
Date: Tue, 28 Oct 1997 00:12:06 -0700 (MST)
Cc: flexfax@sgi.com
Read-Receipt-To: steve@genie96.com
Sender: owner-flexfax@celestial.com

> 
> 
> I hope the following appraoch will meet everybody's concerns.
> I will write the rpm documentation as follows:
> 
> 	Once the rpm is installed, run 'faxsetup'. 'faxsetup'
> 	invokes another script in turn 'faxaddmodem'. If you don't 
> 	experience problems like modem hanging, you most probably
> 	have a close to working version. Check if you are able to send
> 	and receive faxes. If there is any problem, time to look at
> 	the config files. There are two of them
> 	/var/spool/hyla/etc/config and
> 	/var/spool/hyla/etc/config.modem. Do a 'man config' and get
> 	an overview of all the parameters that can go in there. Try 
> 	changing them.
> 
> 	If your modem hangs during 'faxaddmodem', you might want 
> 	to attempt a manual setup. copy the files
> 	config, config.modem, setup, and setup.modem from
> 	<to be specified directory> to /var/spool/hyla/etc.
> 	Look at /var/spool/hyla/config and see what looks closest
> 	to your modem. Adjust the config parameters through trial 
> 	and error.
> 

Hi,

The above looks good.  Just a "gentle" suggestion, from my AIX install
notes:

- If faxaddmodem ( or faxsetup ) hanges at "
  Probing for best speed to talk to modem:

  Make sure that your modem has verbose result codes enabled.  If that is not
  sufficient to make it work, try setting Carrier Dectect to be forced on.
  ( Typically &C0 ).  This is not necessary under all versions of AIX, but
  I did find it necessary some times.


This might help solve some problems.

-- 
	Steve Williams, Calgary, Alberta, Canada
	Genie Computer Systems Inc.
	steve@genie96.com

"A man doesn't begin to attain wisdom until he recognizes that he is 
 no longer indespensable."
- Admiral Richard E. Byrd ( 1888-1957 )

From owner-flexfax@celestial.com  Wed Feb 17 13:55:47 1999
From: Hong Kong Observatory <d1swdev@asiaonline.net>
Subject: flexfax: Re: fix for "Can not open document"
To: guru@SOFTCON.de
Date: Tue, 28 Oct 1997 17:14:15 +0800 (HKT)
Cc: flexfax@sgi.com, I.Senior@bom.gov.au, john@BlueSkyTours.COM
Sender: owner-flexfax@celestial.com

Hi Matthias,

I just download Hylafax V4.0pl1 into my IBM PowerPC running AIX 4.2 with
gcc 2.7.2 and ghostscript 4.02.

I compiled and installed the package without any problem (without your
patch). But I cannot reproduce the 'Cannot open document file' problem.

I issued the command : sendfax -n -d 1111 -d 2222 doc.ps
and made the line 1111 busy. I found that the server did the  conversion
ps2fax twice (once for each FAX). Therefore there is no "cannot open
document file" problem. But I think this is another problem because
the conversion is done once for each destination, this will waste a lot
of CPU resource ! Do you think that I have made any mistake in my
installation ?

Also from the 'syslog' file, I found an error from faxgetty every time
before each dispatch :
Faxgetty [12345]: ANSWER: Can not lock modem device.
but the document was dispatched without errors. Does this imply any
potential porblem ?

Thank you .

K.S.Wat SSA(D)12
Computer Division,
Hong Kong Observatory.

> 
> Hi,
> 
> attached below is what I think should be the fix for the
> "Can not open document problem". Please run tests for this.
> I'm still working on some HTML && doc changes and RCS freezing
> and could release the v4,0pl2 beta if this patch fixes the problem.
> 
> 	matthias
> 
> 
> *** faxSendApp.c++.orig	Sun Oct 19 10:02:37 1997
> --- faxSendApp.c++	Sun Oct 26 07:27:19 1997
> ***************
> *** 251,259 ****
>   {
>       FaxSendInfo si(req.requests[fi].item, req.commid, req.npages,
>   	getFileTransferTime(), getClientParams());
> -     sendJobStatus(req.jobid, "D%s", (const char*) si.encode());
>   
>       FaxServer::notifyDocumentSent(req, fi);
>   }
>   
>   /*
> --- 251,262 ----
>   {
>       FaxSendInfo si(req.requests[fi].item, req.commid, req.npages,
>   	getFileTransferTime(), getClientParams());
>   
>       FaxServer::notifyDocumentSent(req, fi);
> + 
> +     // NB: there is a racing with the scheduler and we should delay
> +     //     the FIFO message to scheduler until we renamed the document
> +     sendJobStatus(req.jobid, "D%s", (const char*) si.encode());
>   }
>   
>   /*
> -- 
> firm: matthias.apitz@sisis.de [voc:+49 89 61308 351, fax: +49 89 61308 188]
> priv: guru@thias.muc.de
>  WWW: http://www.sisis.de/~guru/ OR http://www.guug.de/GUUG/firmen/apitz/
> 
> "We don't care.  We don't have to.  You'll buy whatever we ship, so why bother?
>  We're Microsoft." 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Re: fix for "Can not open document"
To: d1swdev@asiaonline.net (Hong Kong Observatory)
Date: Tue, 28 Oct 1997 13:25:04 +0100 (MEZ)
Cc: flexfax@sgi.com, I.Senior@bom.gov.au, john@BlueSkyTours.COM
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Hong Kong Observatory wrote:
    
    Hi Matthias,
    
    I just download Hylafax V4.0pl1 into my IBM PowerPC running AIX 4.2 with
    gcc 2.7.2 and ghostscript 4.02.
    
    I compiled and installed the package without any problem (without your
    patch). But I cannot reproduce the 'Cannot open document file' problem.
    
    I issued the command : sendfax -n -d 1111 -d 2222 doc.ps
    and made the line 1111 busy. I found that the server did the  conversion
    ps2fax twice (once for each FAX). Therefore there is no "cannot open
    document file" problem. But I think this is another problem because
    the conversion is done once for each destination, this will waste a lot
    of CPU resource ! Do you think that I have made any mistake in my
    installation ?

No. The racing between faxq(1M) and faxsend(1M) (or call it a bug)
will not show up on each system nor in each situation. With the
unpatched code of v4.0pl1 it's just a gamble which part wins.
    
    Also from the 'syslog' file, I found an error from faxgetty every time
    before each dispatch :
    Faxgetty [12345]: ANSWER: Can not lock modem device.
    but the document was dispatched without errors. Does this imply any
    potential porblem ?

No. The faxsend(1M) plays with the modem to send the fax, the modem
answers with something like "OK" and both (faxsend(1M) and faxgetty(1M))
can "see" (with poll(2) or select(3C)) that the modem says something
in response to the AT-cmd (sent down by faxsend(1M)), but because
the faxsend(1M) has the lock for the device, faxgetty(1M) goes out
of the way stumbling about it with "Can not lock modem device"
and waits for the lock to be cleared. That's all.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Hong Kong Observatory <d1swdev@asiaonline.net>
Subject: flexfax: Re: fix for "Can not open document"
To: guru@SOFTCON.de
Date: Wed, 29 Oct 1997 08:48:38 +0800 (HKT)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

Hi Matthias,

Please tell how to test your patch then ? Also how to fix the duplicated
document conversion problem (this would waste lots of CPU resource if 
huge document was sent to different destinations) ?

> 
> Hong Kong Observatory wrote:
>     
>     Hi Matthias,
>     
>     I just download Hylafax V4.0pl1 into my IBM PowerPC running AIX 4.2 with
>     gcc 2.7.2 and ghostscript 4.02.
>     
>     I compiled and installed the package without any problem (without your
>     patch). But I cannot reproduce the 'Cannot open document file' problem.
>     
>     I issued the command : sendfax -n -d 1111 -d 2222 doc.ps
>     and made the line 1111 busy. I found that the server did the  conversion
>     ps2fax twice (once for each FAX). Therefore there is no "cannot open
>     document file" problem. But I think this is another problem because
>     the conversion is done once for each destination, this will waste a lot
>     of CPU resource ! Do you think that I have made any mistake in my
>     installation ?
> 
> No. The racing between faxq(1M) and faxsend(1M) (or call it a bug)
> will not show up on each system nor in each situation. With the
> unpatched code of v4.0pl1 it's just a gamble which part wins.
>     
>     Also from the 'syslog' file, I found an error from faxgetty every time
>     before each dispatch :
>     Faxgetty [12345]: ANSWER: Can not lock modem device.
>     but the document was dispatched without errors. Does this imply any
>     potential porblem ?
> 
> No. The faxsend(1M) plays with the modem to send the fax, the modem
> answers with something like "OK" and both (faxsend(1M) and faxgetty(1M))
> can "see" (with poll(2) or select(3C)) that the modem says something
> in response to the AT-cmd (sent down by faxsend(1M)), but because
> the faxsend(1M) has the lock for the device, faxgetty(1M) goes out
> of the way stumbling about it with "Can not lock modem device"
> and waits for the lock to be cleared. That's all.
> 
> 	matthias
> 

K.S.Wat SSA(D)12
Computer Division,
Hong Kong Observatory.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: flexfax: Re: fix for "Can not open document"
To: d1swdev@asiaonline.net (Hong Kong Observatory)
Date: Wed, 29 Oct 1997 07:59:58 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Hong Kong Observatory wrote:
    
    Hi Matthias,
    
    Please tell how to test your patch then ? Also how to fix the duplicated
    document conversion problem (this would waste lots of CPU resource if 
    huge document was sent to different destinations) ?

How to reproduce this problem was exactly described by
Ian Senior in the mailing-list Mar 5, 97 (please pull it
from the archives at http://www.vix.com/hylafax/),

If the same document is formatted twice check als why by
enabling the serverTrace. There are reasons why the
document must be converted twice (e.g. different capabiltities
of the destinations).

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Hong Kong Observatory <d1swdev@asiaonline.net>
Subject: flexfax: Re: fix for "Can not open document"
To: Matthias.Apitz@SOFTCON.de
Date: Fri, 31 Oct 1997 15:12:59 +0800 (HKT)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

Hi Matthias,

I followed the 'Ian Senior's mail' for the test but until lastly all fax
jobs sent, all files remain stayed in directory 'docq' ! I just installed
HylaFAX in my system for a week and all testing files used within these
few days still remain in 'docq'. How will these files be removed ? Do
you (or Ian) mean that the 'Cannot open document file' problem is caused
by running 'faxqclean' ?

I'm so sorry that I still cannot produce the problem !

> 
> How to reproduce this problem was exactly described by
> Ian Senior in the mailing-list Mar 5, 97 (please pull it
> from the archives at http://www.vix.com/hylafax/),
> 
> 	matthias
> 

K.S.Wat SSA(D)12
Computer Division,
Hong Kong Observatory.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Wed, 29 Oct 1997 03:08:32 -0600
To: David Woolley <david@djwhome.demon.co.uk>
From: "Scott R. Keszler" <keszler@rrnet.com>
Subject: flexfax: Re: (Missing Subject)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

At 08:20 AM 10/28/97 +0000, David Woolley wrote:
>> Running hylafax-v4.0pl1 on Linux 1.3-3#17 on a Cyrix 6x86L (P166, not MMX)
>                                   ^^^^^^^^

Yeouch - I didn't even notice that.  I'm running 2.0.29; I must have
slipped cutting the string out of the xterm.

I've been beating my head against the wall on this - from two different
mfgrs/models of fax machine I get the following:

Oct 28 16:53:17.36: [ 2890]: --> [17:+FPTS: 1,2219,0,0]
Oct 28 16:53:18.70: [ 2890]: --> [7:+FET: 2]
Oct 28 16:53:18.70: [ 2890]: RECV recv EOP (no more pages or documents)
Oct 28 16:53:18.81: [ 2890]: --> [2:OK]
Oct 28 16:53:18.81: [ 2890]: RECV send RTN (retrain negative)
Oct 28 16:53:18.81: [ 2890]: <-- [7:AT+FDR\r]
Oct 28 16:53:21.53: [ 2890]: --> [8:+FHNG: 0]

but receiving from a fax modem (motorola class 2 on an AIX box running
proprietary fax software - imaging system) I get:

Oct 28 17:52:37.62: [ 2890]: --> [17:+FPTS: 1,2219,0,0]
Oct 28 17:52:38.84: [ 2890]: --> [7:+FET: 2]
Oct 28 17:52:38.84: [ 2890]: RECV recv EOP (no more pages or documents)
Oct 28 17:52:38.84: [ 2890]: --> [2:OK]
Oct 28 17:52:38.84: [ 2890]: RECV send MCF (message confirmation)
Oct 28 17:52:38.84: [ 2890]: <-- [7:AT+FDR\r]
Oct 28 17:52:41.47: [ 2890]: --> [8:+FHNG: 0]

When receiving from a fax machine, hylaFAX sends a retrain request but
the fax machine hangs up.  When receiving from a fax modem, hylaFAX
sends a message confirmation and accepts the page. I've watched as the
fax arrives from both fax machines - the tif is created in recvq, grows
to the correct size, then gets deleted when the fax machine hangs up.

I haven't been able to find any reason why hylaFAX is sending a retrain
request - no receive errors appear the the log - or why the fax machine
is ignoring the request.  I've tried adding the config line shown on the
web page for class 2 modems to no effect. I've set the session tracing
to 131071 and not seen anything indicating the problem.  I've set the
fax modem's port speed from 38400 down to 9600.  No matter what I do,
it sends a retrain request and gets hung up on by the fax machine.

Ideas?



-- 
Dead Hard Drives make great clocks: http://rrnet.com/~keszler/hddclock.html

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Comments: Authenticated sender is <spine001@gold.tc.umn.edu>
From: Julio C Spinelli <spine001@gold.tc.umn.edu>
To: "Scott R. Keszler" <keszler@rrnet.com>
Date: Wed, 29 Oct 1997 09:52:57 +0000
Subject: Re: flexfax: Re: (Missing Subject)
CC: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

No clue is this has anything to do, but check on the copy quality 
checking response that your modem provides, some modems say that they 
do copy quality checking but they don't. You can always overrule the 
modem response in the config.xxx file. You may also want to set a low 
percentage of good lines and a high number of consecutive bad lines 
as acceptable. And see what happens...

Julio

Note: may be forcing a lower baud rate will help.

> Date:          Wed, 29 Oct 1997 03:08:32 -0600
> To:            David Woolley <david@djwhome.demon.co.uk>
> From:          "Scott R. Keszler" <keszler@rrnet.com>
> Subject:       flexfax: Re: (Missing Subject)
> Cc:            flexfax@sgi.com

> At 08:20 AM 10/28/97 +0000, David Woolley wrote:
> >> Running hylafax-v4.0pl1 on Linux 1.3-3#17 on a Cyrix 6x86L (P166, not MMX)
> >                                   ^^^^^^^^
> 
> Yeouch - I didn't even notice that.  I'm running 2.0.29; I must have
> slipped cutting the string out of the xterm.
> 
> I've been beating my head against the wall on this - from two different
> mfgrs/models of fax machine I get the following:
> 
> Oct 28 16:53:17.36: [ 2890]: --> [17:+FPTS: 1,2219,0,0]
> Oct 28 16:53:18.70: [ 2890]: --> [7:+FET: 2]
> Oct 28 16:53:18.70: [ 2890]: RECV recv EOP (no more pages or documents)
> Oct 28 16:53:18.81: [ 2890]: --> [2:OK]
> Oct 28 16:53:18.81: [ 2890]: RECV send RTN (retrain negative)
> Oct 28 16:53:18.81: [ 2890]: <-- [7:AT+FDR\r]
> Oct 28 16:53:21.53: [ 2890]: --> [8:+FHNG: 0]
> 
> but receiving from a fax modem (motorola class 2 on an AIX box running
> proprietary fax software - imaging system) I get:
> 
> Oct 28 17:52:37.62: [ 2890]: --> [17:+FPTS: 1,2219,0,0]
> Oct 28 17:52:38.84: [ 2890]: --> [7:+FET: 2]
> Oct 28 17:52:38.84: [ 2890]: RECV recv EOP (no more pages or documents)
> Oct 28 17:52:38.84: [ 2890]: --> [2:OK]
> Oct 28 17:52:38.84: [ 2890]: RECV send MCF (message confirmation)
> Oct 28 17:52:38.84: [ 2890]: <-- [7:AT+FDR\r]
> Oct 28 17:52:41.47: [ 2890]: --> [8:+FHNG: 0]
> 
> When receiving from a fax machine, hylaFAX sends a retrain request but
> the fax machine hangs up.  When receiving from a fax modem, hylaFAX
> sends a message confirmation and accepts the page. I've watched as the
> fax arrives from both fax machines - the tif is created in recvq, grows
> to the correct size, then gets deleted when the fax machine hangs up.
> 
> I haven't been able to find any reason why hylaFAX is sending a retrain
> request - no receive errors appear the the log - or why the fax machine
> is ignoring the request.  I've tried adding the config line shown on the
> web page for class 2 modems to no effect. I've set the session tracing
> to 131071 and not seen anything indicating the problem.  I've set the
> fax modem's port speed from 38400 down to 9600.  No matter what I do,
> it sends a retrain request and gets hung up on by the fax machine.
> 
> Ideas?
> 
> 
> 
> -- 
> Dead Hard Drives make great clocks: http://rrnet.com/~keszler/hddclock.html
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Tue, 28 Oct 1997 14:34:42 -0500
From: Ramana Juvvadi <juvvadi@lekha.org>
To: flexfax@sgi.com
Subject: flexfax: Splitting the rpm into client, server, and lib packages
Sender: owner-flexfax@celestial.com


Finally, I managed to build an rpm with shared libraries
libfaxserver.so and libfaxutil.so. But libregex still static. 
I will release it as soon as some other minor details
are taken care of.

I am wondering if it makes sense to split the Hylafax package
into server, client, and library pieces. The idea is, machines
that need to run as only clients don't have to install the
server packages. If so, which files go into the server and which 
files go into the client package.

Ramana

 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Tue, 28 Oct 1997 16:34:29 -0500
From: Ramana Juvvadi <juvvadi@lekha.org>
To: "Eric M. Boehm" <boehm@engineous.com>
Cc: flexfax@sgi.com (HylaFAX Mailing List)
Subject: flexfax: Problems receiving faxes
Sender: owner-flexfax@celestial.com

Eric M. Boehm writes:

 > Oct 28 13:33:20.09: [26001]: RECV: send trigger 021

I have a Cardinal MVPV34I. Setting send trigger to 022 
solved the problem. If the chipset is similar it might work.
Here is what I have in my config file.

ModemSendFillOrder:     LSB2MSB         # as expected
ModemRecvFillOrder:     MSB2LSB         # opposite of what makes sense

Class2RecvDataTrigger:  "\022"          # character sent to modem to start recv


Ramana

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: "Eric M. Boehm" <boehm@engineous.com>
Subject: Re: flexfax: Problems receiving faxes
To: juvvadi@lekha.org (Ramana Juvvadi)
Date: Tue, 28 Oct 1997 15:53:15 -0500 (EST)
Cc: flexfax@sgi.com (HylaFAX Mailing List)
Reply-To: "Eric M. Boehm" <boehm@engineous.com>
X-PGP-Fingerprint: CA 5F 8B 86 57 BF 60 74  A3 86 62 E7 79 BA 98 0E
Sender: owner-flexfax@celestial.com

>>>>> "Ramana" == Ramana Juvvadi <juvvadi@lekha.org> writes:

    Ramana> I have a Cardinal MVPV34I. Setting send trigger to 022 solved the
    Ramana> problem. If the chipset is similar it might work.  Here is what I
    Ramana> have in my config file.

    Ramana> ModemSendFillOrder: LSB2MSB # as expected
    Ramana> ModemRecvFillOrder:MSB2LSB # opposite of what makes sense
    Ramana> Class2RecvDataTrigger: "\022" # character sent to modem to start recv

Well, that seems to work. I looked at the
received fax it appears to be all there. I am a little worried though -- the
sessions trace indicates that there was a problem

Oct 28 15:39:36.39: [26541]: RECV: begin page
Oct 28 15:39:36.40: [26541]: MODEM set XON/XOFF/FLUSH: input ignored, output gen
erated
Oct 28 15:39:36.40: [26541]: RECV: send trigger 022
Oct 28 15:39:36.40: [26541]: <-- data [1]
Oct 28 15:40:11.50: [26541]: RECV/CQ: Bad 1D pixel count, row 2122, got 0, expec
ted 1728
Oct 28 15:40:11.50: [26541]: RECV/CQ: Bad 1D pixel count, row 2123, got 0, expec
ted 1728
Oct 28 15:40:11.51: [26541]: RECV/CQ: Bad 1D pixel count, row 2124, got 0, expec
ted 1728
Oct 28 15:40:11.51: [26541]: RECV/CQ: Bad 1D pixel count, row 2125, got 0, expec
ted 1728
Oct 28 15:40:11.51: [26541]: RECV/CQ: Bad 1D pixel count, row 2126, got 0, expec
ted 1728
Oct 28 15:40:11.65: [26541]: RECV/CQ: Bad 1D pixel count, row 2127, got 3134, ex
pected 1728
Oct 28 15:40:11.65: [26541]: RECV/CQ: Adjusting for trailing noise (6 run)
Oct 28 15:40:11.65: [26541]: RECV: 2122 total lines, 0 bad lines, 0 consecutive
bad lines
Oct 28 15:40:11.65: [26541]: MODEM set XON/XOFF/DRAIN: input interpreted, output
 disabled
Oct 28 15:40:11.66: [26541]: --> [17:+FPTS: 2,2219,0,0]
Oct 28 15:40:12.89: [26541]: --> [7:+FET: 2]

I see similar results for a multi-page fax -- should I be concerned?

-- 
Eric M. Boehm                                 boehm@engineous.com
Senior System Analyst                         Engineous Software, Inc.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
X-Authentication-Warning: tui.pinnacle.co.nz: jonc owned process doing -bs
Date: Wed, 29 Oct 1997 13:42:15 +1300 (NZDT)
From: Jonathan Chen <jonc@pinnacle.co.nz>
To: "Eric M. Boehm" <boehm@engineous.com>
cc: HylaFAX Mailing List <flexfax@sgi.com>
Subject: Re: flexfax: Problems receiving faxes
Sender: owner-flexfax@celestial.com

On Tue, 28 Oct 1997, Eric M. Boehm wrote:

> I have hylfax v4.0pl1 set up on a Sparc 5, Solaris 2.5.1 with two Cardinal
> MVP288XF modems. The modems have had their firmware update so that they are
> really 33.6 modems. They use a Rockwell RC288dpi.

Try addiing/modifying the following lines:

    ModemRate:          38400           # locked for fax sending & receiving
    ModemFlowControl:   rtscts          # default

    ...

    #
    # Chip locks the line rate and switches to xon/xoff flow
    # control when receiving facsimile--we must follow suit.
    #
    ModemAnswerFaxBeginCmd: "<19200><xon>"  # switch line rate & flow control

We're using RC288DPi based modems as well, and it works fine with
ours.

Cheers.
--
Jonathan Chen <jonc@pinnacle.co.nz>                 Once is dumb luck.
                                                 Twice is coincidence.
             Three times and Somebody Is Trying To Tell You Something.


From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Wed, 22 Oct 1997 13:11:09 GMT
From: "J.B. LaRue" <jblarue@sparrow.mila-inc.com>
To: flexfax@sgi.com
Subject: flexfax: ZyXEL config
Cc: jb.larue@mila-inc.com
Sender: owner-flexfax@celestial.com



Well, after quite some time of fudging the config for a ZyXEL 1496E, I found
that by lowering the baud rate to 19200 in both the ModemRate as well as
the S18 register, I am able to run with substantially less errors. Actually,
I'm running 99.8% error free. We're receiving between 100-200 faxes a day,
most of large quantities (40+ pages). I'll include my config below. There are
16 modems in a rack...


#
CountryCode:		1
AreaCode:		206
FAXNumber:		+1.222.222.2222
LongDistancePrefix:	1
InternationalPrefix:	011
DialStringRules:	"etc/dialrules"
ServerTracing:		1
SessionTracing:		11
RecvFileMode:		0600
LogFileMode:		0600
DeviceMode:		0600
RingsBeforeAnswer:	1
SpeakerVolume:		medium
GettyArgs:		"-h %l dx_%s"
LocalIdentifier:	"MAIN, Inc."
TagLineFont:		etc/lutRS18.pcf
TagLineFormat:		"From %%l|%c|Page %%p of %%t"
NoCarrierRetrys:	3
MaxRecvPages:		75
#
#
# Modem-related stuff: should reflect modem command interface
# and hardware connection/cabling (e.g. flow control).
#
ModemType:		Class2.0	# use class 2.0 interface
# ModemRate:		38400		# 38.4 works fine
ModemRate:		19200		# 19.2 works fine
ModemFlowControl:	rtscts		# default
#
ModemNoFlowCmd:		AT&H0		# disable flow control
ModemHardFlowCmd:	AT&H3		# hardware flow control
ModemSoftFlowCmd:	AT&H4		# software flow control
ModemSetupDTRCmd:	AT&D2		# DTR off causes modem to hang up
ModemSetupDCDCmd:	AT&C1		# DCD tracks carrier
#
ModemSetupAACmd:	AT+FAA=1	# enable in current class
#
# Additional reset commands:
#
# &B1	  DTE-DCE rate is fixed at DTE setting
# &N0	  Auto-negotiate highest possible DCE-DCE link rate
# &S0	  DSR always on
# *F0	  Deny remote configuration
# S18=3	  Receive at 19200
# S38.3=1 DCD on/off sequence follows UNIX standard; also
#	  fix receiving baud rate at S18 value
# S39=0	  (avoid Class 2 compatibility hacks)
# 
ModemResetCmds:		AT&B1&N0&S0*F0S18=3S38.3=1S39=0
#
ModemDialCmd:		ATDT%s		# no '@' 'cuz then busy not recognized
NoCarrierRetrys:	3		# retry 3 times on no carrier
#
ModemSetVolumeCmd:	"ATM0 ATM1L1 ATM1L3 ATM1L5 ATM1L7"
#
# Caller ID configuration setup.
#
#QualifyCID:		etc/cid		# CID access control list file
#CIDNumber:		"CALLER NUMBER: " # pattern string for phone number info
#CIDName:		"CALLER NAME: "	# pattern string for identity info
#
PagerSetupCmds:		AT&K0&N15	# use V.22 at 1200 bps (PageNet)
#
# Rev 6.1x firmware have a bug in the ECM support so
# explicitly disable it's use.  To re-enable its use just
# comment out the following line.
#
Class2DCCQueryCmd:	"!(0,1),(0-5),(0-4),(0-2),(0,1),(0),(0),(0-7)"

Hope this helps anyone...

jb

<><><><><><><><><><><><><><><><><><><><><><><><><><><><>
Does "Systems Administration" also mean phone "systems"?
<><><><><><><><><><><><><><><><><><><><><><><><><><><><>

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Wed, 22 Oct 1997 11:37:47 -0500 (CDT)
From: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
To: "Paul P.H. Wilson" <wilson@maple.fzk.de>
cc: HylaFAX Mailing List <flexfax@sgi.com>
Subject: Re: flexfax: Communications Problems
Sender: owner-flexfax@celestial.com

Most Rockwell modems switch to software handshaking during fax
transmission and they also lock the baud rate at 19200. Is this switch
correctly programmed in your config file? Check some of the Rockwell
config files that come with the package,

Julio

On Wed, 22 Oct 1997, Paul P.H. Wilson wrote:

> Hello again,
> 
> I originally wrote to the list about 3 weeks ago with
> communications problems when using my Faxmodem to answer
> calls with faxgetty.
> 
> The problems are persisting (I have only just had time to
> resume testing), and are completely eluding me.  If anyone
> can help me understand the following, I would be grateful.
> I have experienced this problem from two different fax
> machines and one other faxmodem.  I can send faxes without
> any trouble.
> 
> Thanks in advance!
> 
> ServerLog:
> 
> FaxGetty: OPEN /dev/cua1
> FaxGetty: MODEM set DTR OFF
> FaxGetty: DELAY 2600 ms
> FaxGetty: MODEM set DTR ON
> FaxGetty: MODEM set baud rate: 38400 baud, input flow
> XON/XOFF, output flow XON/XOFF
> FaxGetty: MODEM flush i/o
> FaxGetty: <-- [14:ATE0V1Q0S0=0H\r]
> FaxGetty: --> [2:OK]
> FaxGetty: <-- [21:ATS8=2S7=60&K4&D2&C1\r]
> FaxGetty: --> [2:OK]
> FaxGetty: <-- [12:AT+FCLASS=?\r]
> FaxGetty: --> [5:0,1,2]
> FaxGetty: --> [2:OK]
> FaxGetty: MODEM Supports "Data"
> FaxGetty: MODEM Supports "Class 1"
> FaxGetty: MODEM Supports "Class 2"
> FaxGetty: <-- [12:AT+FCLASS=2\r]
> FaxGetty: --> [2:OK]
> FaxGetty: <-- [9:AT+FMFR?\r]
> FaxGetty: --> [20:ROCKWELL            ]
> FaxGetty: --> [2:OK]
> FaxGetty: MODEM: Mfr ROCKWELL
> FaxGetty: <-- [9:AT+FMDL?\r]
> FaxGetty: --> [20:RC32ACL             ]
> FaxGetty: --> [2:OK]
> FaxGetty: <-- [9:AT+FREV?\r]
> FaxGetty: --> [18:CES-04A GER 940829]
> FaxGetty: --> [2:OK]
> FaxGetty: MODEM: Model RC32ACL
> FaxGetty: MODEM: Revision CES-04A GER 940829
> FaxGetty: <-- [10:AT+FDCC=?\r]
> FaxGetty: --> [39:(0,1),(0-5),(0-2),(0-2),(0,1),0,0,(0-7)]
> FaxGetty: --> [2:OK]
> FaxGetty: MODEM Supports 3.85 line/mm
> FaxGetty: MODEM Supports 7.7 line/mm
> FaxGetty: MODEM Supports 2400 bit/s
> FaxGetty: MODEM Supports 4800 bit/s
> FaxGetty: MODEM Supports 7200 bit/s
> FaxGetty: MODEM Supports 9600 bit/s
> FaxGetty: MODEM Supports 12000 bit/s
> FaxGetty: MODEM Supports 14400 bit/s
> FaxGetty: MODEM Supports page width 1728 pixels in 215 mm
> FaxGetty: MODEM Supports page width 2048 pixels in 255 mm
> FaxGetty: MODEM Supports page width 2432 pixels in 303 mm
> FaxGetty: MODEM Supports A4 page length (297 mm)
> FaxGetty: MODEM Supports B4 page length (364 mm)
> FaxGetty: MODEM Supports unlimited page length
> FaxGetty: MODEM Supports 1-D MR
> FaxGetty: MODEM Supports 2-D MR
> FaxGetty: MODEM Supports 0 ms/scanline
> FaxGetty: MODEM Supports 5 ms/scanline
> FaxGetty: MODEM Supports 10 ms, 5 ms/scanline
> FaxGetty: MODEM Supports 10 ms/scanline
> FaxGetty: MODEM Supports 20 ms, 10 ms/scanline
> FaxGetty: MODEM Supports 20 ms/scanline
> FaxGetty: MODEM Supports 40 ms, 20 ms/scanline
> FaxGetty: MODEM Supports 40 ms/scanline
> FaxGetty: <-- [9:AT+FCQ=?\r]
> FaxGetty: --> [1:0]
> FaxGetty: --> [2:OK]
> FaxGetty: MODEM Supports no copy quality checking
> FaxGetty: <-- [9:AT+FAP=?\r]
> FaxGetty: --> [5:ERROR]
> FaxGetty: <-- [10:AT+FSPL=?\r]
> FaxGetty: --> [1:0]
> FaxGetty: --> [2:OK]
> FaxGetty: <-- [12:AT+FCLASS=2\r]
> FaxGetty: --> [2:OK]
> FaxGetty: <-- [10:AT+FTBC=0\r]
> FaxGetty: --> [2:OK]
> FaxGetty: <-- [10:AT+FBOR=0\r]
> FaxGetty: --> [2:OK]
> FaxGetty: <-- [13:AT+FPHCTO=30\r]
> FaxGetty: --> [2:OK]
> FaxGetty: <-- [10:AT+FBUG=1\r]
> FaxGetty: --> [5:ERROR]
> FaxGetty: MODEM Command error
> FaxGetty: <-- [24:AT+FDCC=1,5,2,2,1,0,0,0\r]
> FaxGetty: --> [2:OK]
> FaxGetty: MODEM ROCKWELL RC32ACL/CES-04A GER 940829
> FaxGetty: <-- [7:ATL2M1\r]
> FaxGetty: --> [2:OK]
> FaxGetty: <-- [9:AT+FCR=1\r]
> FaxGetty: --> [2:OK]
> FaxGetty: <-- [31:AT+FLID="INR Fusion Technolog"\r]
> FaxGetty: --> [2:OK]
> FaxGetty: <-- [19:AT+FCLASS=0;+FAA=1\r]
> FaxGetty: --> [2:OK]
> FaxGetty: MODEM input buffering disabled
> FaxGetty: MODEM input buffering enabled
> FaxGetty: --> [4:RING]
> FaxGetty: --> [4:RING]
> FaxGetty: MODEM input buffering enabled
> FaxGetty: ANSWER: FAX CONNECTION
> FaxGetty: <-- [4:ATH\r]
> FaxGetty: --> [2:OK]
> 
> SessionLog of failed session:
> 
> : SESSION BEGIN 00000014 497247822047
> : <-- [4:ATA\r]
> : --> [3:FAX]
> : --> [5:+FCON]
> : ANSWER: FAX CONNECTION
> : STATE CHANGE: ANSWERING -> RECEIVING
> : MODEM input buffering enabled
> : RECV FAX: begin
> : MODEM TIMEOUT: reading line from modem
> : REMOTE HANGUP: Unspecified Receive Phase B error (code 70)
> 
> : RECV FAX: Unspecified Receive Phase B error
> : RECV FAX: end
> : SESSION END
> 
> --
> Paul Wilson                wilson@maple.fzk.de
>         http://maple.fzk.de/~wilson/
> **********************************************
> *       My Canada Includes Quebec            *
> * Another Environmentalist for Nuclear Power *
> **********************************************
> 
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: flexfax: Re: FAQ#85
To: wilson@maple.fzk.de (Paul P.H. Wilson)
Date: Wed, 22 Oct 1997 18:41:03 +0200 (MSZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Paul P.H. Wilson wrote:
    
    Hello HylaFAX maintainers,

    I have finally got HylaFAX working wonderfully on my Linux
    box (I had to find the little note about setting the
    ModemRate to 19200 for my apparently Supra like modem :).
    
    I am interested in doing something like the original asker
    or Q85 had in mind.  Namely, I would like to have multiple
    telephone numbers ring on the same line/modem and then have
    the faxes sent by email to different users, dependent on the
    original number dialed.
    
    I have looked up DID in the glossary which you suggested,
    but it is still not clear to me whether or not (and if so
    how) HylaFAX supports DID?  How is this originally dialed
    number passed to HylaFAX?  How is it possible to tell if the
    faxmodem itself supports DID (as the glossary suggests is
    possible).

HylaFAX v4.0pl1 doesn't support DID (see also the HTML-page
html/setup-advanced.html in the source distribution).

With DSS1 ISDN (very common in Europe) and a ZyXEL Elite 2864I
modem/TA hybrid (which acts as a ISDN TA *and* as a fax/data
modem in one device) one could do the following trick to get
nearly what you want:

- order DSS1 ISDN and as much MSNs (Multiple Subscriber Numbers
  as you need from Telco;
- organize with the Telco that all start with the same digits
  and are different in the last digits (e.g. 123456-100, 123456-101
  123456-102 ....)
- all calls to this number will be delivered to the same ISDN So bus
  and your ZyXEL Elite 2864I will "see" all these calls;
- configure the ZyXEL Elite 2864I to pick-up incoming calls for
  numbers containing the string "123456" as fax/data-calls
  and let faxgetty(1M) answer a call on the first(!) ring;
- each session-log will contain the dialed number as in this example:

  Oct 22 20:20:06.86: [  672]: SESSION BEGIN 00000033 498943651320
  Oct 22 20:20:06.97: [  672]: <-- [4:ATA\r]
  Oct 22 20:20:07.10: [  672]: --> [25:FM:8961308194 TO:123456100]
                                                     ^^^^^^^^^^^^
  Oct 22 20:20:07.50: [  672]: --> [30:CONNECT]
  Oct 22 20:20:07.50: [  672]: ANSWER: FAX CONNECTION
	....

- you may change the faxrcvd(1M) script to grep the number "123456100"
  from the session-log and use it for e-mail routing;
  (if you don't answer on first ring, the "FM:8961308194 TO:123456100"
  message is in the server trace and the logic is more complex);

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 23 Oct 1997 10:37:39 +0200
From: Uri Shkolnik <noaml@netvision.net.il>
Reply-To: uri@harmonic.co.il
To: HylaFax <flexfax@sgi.com>
Subject: flexfax: fax2tiff - no page breaks anymore
Sender: owner-flexfax@celestial.com

Hi,

Till now I had used fax2ps, with postscript viewer installed on the
users PC.
Now I want to use tiff format.
I tries to run the command:

fax2tiff -4 -s -o  <for_user.tif> <incoming_fax_format.tif>

And view it with "Imaging" for Win** (by Wang)

The output is good and clear, but with one problem, there is no page
breaks any more. A multipages incoming fax look as continues image...

What can I do?


Uri

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: kevin@meso.com
To: flexfax@sgi.com
Cc: kevin@meso.com
Subject: flexfax: Headers and ASCII-only files 
Date: Thu, 23 Oct 97 10:46:52 -0400
X-Mts: smtp
Sender: owner-flexfax@celestial.com

Hi,

Is there a way to control the amount of whitespace that appears
between the header and the body when sending an ASCII file?  We
are currently using a commercial product to send out our faxes 
and its amount of whitespace is small enough that our files
fit on one page.  When using HylaFAX, it now takes two pages.

Any info greatly appreciated.  Thanks!

Kevin Tyle
MESO, Inc.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: kevin@meso.com
To: flexfax@sgi.com
Cc: kevin@meso.com
Subject: flexfax: Received Fax question 
Date: Fri, 24 Oct 97 10:58:44 -0400
X-Mts: smtp
Sender: owner-flexfax@celestial.com

Hi folks,

I'm using a MultiTech Systems MultiModem, MT2834ZPX.
The config file was built using faxaddmodem without
straying from the defaults.  However, when viewing incoming
text faxes (which were sent as Postscript files) with
Ghostscript (using either fax2ps or fax2tiff), the images
looked distorted ("smudged" is an apropos term).  I edited
the modem config file setting ModemRate down to 9600 and it
looked much better.  I also tried leaving ModemRate at 38400
and set ModemFlowControl to xonxoff instead of rtscts and it
also looked fine.  

Is this an appropriate workaround?  Or have I created potential
new problems?  

"Modem related stuff" included below.  Note that the top
of the config file said that configuration was for
"MT1432, MT224BA, and MT1932ZDX modems."  Are there important
changes that should be made for a MT2834ZPX?

Thanks!

Kevin Tyle
MESO, Inc.

#
#
# Modem-related stuff: should reflect modem command interface
# and hardware connection/cabling (e.g. flow control).
#
ModemType:		Class2		# use class 2 interface
ModemRate:		38400		# lock rate for DCE-DTE communication
#ModemFlowControl:	rtscts		# default
ModemFlowControl:	xonxoff		# default
#
ModemHardFlowCmd:	AT&E4		# hardware flow control
ModemSoftFlowCmd:	AT&E5		# software flow control
ModemSetupDTRCmd:	AT&D3		# setup so DTR drop resets modem
ModemSetupDCDCmd:	AT&C1		# setup so DCD reflects carrier (or not)
ModemSetupAACmd:	AT+FAA=1	# enable adaptive-answer in class 2
#
# NB: some models get confused by the @
#
ModemDialCmd:		ATDT%s@		# T for tone dialing, @ for silence
#
Class2RecvDataTrigger:	"\022"		# character sent to modem to start recv

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: frank@gondor.nikoma.de
Subject: flexfax: infos about sent faxes?
To: flexfax@sgi.com
Date: Fri, 24 Oct 1997 17:45:24 +0200 (CEST)
Sender: owner-flexfax@celestial.com

Hi there!

I'm writing a client program that needs to request
status information (like send status, server dial string and
such) from a Hylafax server to see if a fax has been
successfully sent, even if the fax was submitted three weeks 
ago. This information is saved in the 'doneq' directory, but 
its contents get deleted periodically, together with the according
postscript or tiff files.
The problem is that I only need the status information, but
not the faxes themselves because their total size could in-
crease to gigs of unused data if they aren't deleted. The status 
information I need is saved in etc/xferlog, which cannot be
accessed from a client program.

So here is my question:
Is it possible to keep the status information on the server
accessible to all (which is not a security problem because
you can see status infos about others' faxes when they were
just sent), but to delete the faxes themselves automatically?
(maybe this requires a change in faxcron?)

A better solution would be a mechanism to find out if a fax
was successfully sent without changing the server's configuration
- probably someone knows a way to do this??

thanks in advance,

Frank

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: "Alan Sparks" <asparks@nss.harris.com>
To: <flexfax@sgi.com>
Subject: flexfax: Q: setting the time zone
Date: Fri, 24 Oct 1997 15:43:24 -0700
Sender: owner-flexfax@celestial.com

 All of my faxstat and other reports seem to be showing the time in GMT.  Is
there a setting in the hyla.conf or other file that can force an alternate
TZ setting?

TIA... -Alan

---------
Alan Sparks, IS Engineering Support                   <asparks@harris.com>
Harris Network Support Systems, Camarillo CA    (805) 389-2430

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: infos about sent faxes?
To: frank@gondor.nikoma.de
Date: Sat, 25 Oct 1997 17:43:19 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

frank@gondor.nikoma.de wrote:
    
    Hi there!
    
    I'm writing a client program that needs to request
    status information (like send status, server dial string and
    such) from a Hylafax server to see if a fax has been
    successfully sent, even if the fax was submitted three weeks 
    ago. This information is saved in the 'doneq' directory, but 
    its contents get deleted periodically, together with the according
    postscript or tiff files.
    The problem is that I only need the status information, but
    not the faxes themselves because their total size could in-
    crease to gigs of unused data if they aren't deleted. The status 
    information I need is saved in etc/xferlog, which cannot be
    accessed from a client program.
    
    So here is my question:
    Is it possible to keep the status information on the server
    accessible to all (which is not a security problem because
    you can see status infos about others' faxes when they were
    just sent), but to delete the faxes themselves automatically?
    (maybe this requires a change in faxcron?)

The complete information about a finished job is as long accessible
through the existing client/server protocol as the q#### file
is in the spool/doneq. There is also already a hook (see -a
option) in faxqclean(1M) and in the job file (see "doneop" in
sendq(4F)) to control what should be done with the job by
faxqclean(1M). One solution that comes into my mind is expanding
faxqclean(1M) to let the q#### files in the doneq but remove
the assosiated documents by using the "archive" value of
"doneop" and a new flag in faxqclean(1M). BTW: faxcron(1M) has
nothing to do with the job files or documents.
    
    A better solution would be a mechanism to find out if a fax
    was successfully sent without changing the server's configuration
    - probably someone knows a way to do this??
    
    thanks in advance,

If your client just wants to know if the job was successful or
not there is a notify flag for it (-D of sendfax(1)).

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: fax2tiff - no page breaks anymore
To: uri@harmonic.co.il
Date: Sat, 25 Oct 1997 21:31:46 +0100 (BST)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> 
> fax2tiff -4 -s -o  <for_user.tif> <incoming_fax_format.tif>
> 
> And view it with "Imaging" for Win** (by Wang)

Imaging for Windows, as supplied with NT 4 can directly handle TIFF
files with a Compression tag value of 2, i.e. group 3 fax; you don't
need to convert the files created by Hylafax, although I've only tried
this on a one page junk fax.

This won't necessarily solve your problem though as I don't know how
it would handle multiple page documents.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: DennisLovelady@kemet.com (Dennis Lovelady)
To: flexfax@sgi.com
X-Conversion-ID: <PU-NOTES.2184.878099512.11543>
Organization: KEMET Electronics
Date: Wed, 29 Oct 1997 00:33:03 -0500
Subject: flexfax: Capabilities (mult phone #s, attachments)
Sender: owner-flexfax@celestial.com

Hi, Fax Folks:  :^)

I am new to this list.  I have searched the FAQ and the documentation
.html files, and have not found the answers to some key capability
questions.  If I have simply done a bad job of searching, please let
me know (flame suit is ready).  Thanks.

I have just downloaded HylaFAX 4.0P11, AIX415, GhostScript 3.33.

I would like to know if the following capabilities are available:
   o Fax to list-of-numbers, where the second (and third?) will
     only be used if the first number fails
   o Selection from choice of cover pages made by submitter
   o Include attachments (text and/or graphics)  (How?)

I hope to implement the client as a 'printer,' with a set of 
instructions in the first line(s), including recipient, phone
number (list), list of attachments, notification e-mail address, 
etc.  I don't really know how I will pull that off at this time, 
but that is my hope.

The first implementation, is a Purchase-Order system (canned 
[Oracle Applications]), and that application doesn't have any
sort of 'word processing' interface capabilities, so I fear this
may be problematic.  Anyway, the idea is that the buyer will complete
the purchase order form, and select a print destination which would
be one of:
    Print only
    Print and Fax
    Fax only

Any advice you may have in these regards for an AIX system would be
most appreciated.

Thanks for you time
Dennis Lovelady

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Capabilities (mult phone #s, attachments)
To: DennisLovelady@kemet.com (Dennis Lovelady)
Date: Wed, 29 Oct 1997 10:08:07 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Dennis Lovelady wrote:
    
    Hi, Fax Folks:  :^)
    
    I am new to this list.  I have searched the FAQ and the documentation
    .html files, and have not found the answers to some key capability
    questions.  If I have simply done a bad job of searching, please let
    me know (flame suit is ready).  Thanks.
    
    I have just downloaded HylaFAX 4.0P11, AIX415, GhostScript 3.33.
                                       ^^

BTW: It's not 11 (eleven), it's pl1 (read v4.0pl1 as "version four dot
null patch level one"); I've seen this allready verry often, maybe
because many terminal fonts show a "l" (letter l) similar als "1"
(number one);
    
    I would like to know if the following capabilities are available:
       o Fax to list-of-numbers, where the second (and third?) will
         only be used if the first number fails

No.

       o Selection from choice of cover pages made by submitter

Yes. See man pages of sendfax(1) and faxcover(1);

       o Include attachments (text and/or graphics)  (How?)

No. Preparing the layout of the pages is not the job of HylaFAX.
HylaFAX takes TIFF of PostScript to send it away and has
only some small support for transfering ASCII text (or e-mail
with attachments) into PostScript (see textfmt(1) and faxmail(1));

	matthias
    
    I hope to implement the client as a 'printer,' with a set of 
    instructions in the first line(s), including recipient, phone
    number (list), list of attachments, notification e-mail address, 
    etc.  I don't really know how I will pull that off at this time, 
    but that is my hope.
    
    The first implementation, is a Purchase-Order system (canned 
    [Oracle Applications]), and that application doesn't have any
    sort of 'word processing' interface capabilities, so I fear this
    may be problematic.  Anyway, the idea is that the buyer will complete
    the purchase order form, and select a print destination which would
    be one of:
        Print only
        Print and Fax
        Fax only
    
    Any advice you may have in these regards for an AIX system would be
    most appreciated.
    
    Thanks for you time
    Dennis Lovelady

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Wed, 29 Oct 1997 11:07:48 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: Capabilities (mult phone #s, attachments)
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Wed, 29 Oct 1997 Matthias.Apitz@SOFTCON.de wrote:

> Dennis Lovelady wrote:

>     I would like to know if the following capabilities are available:
>        o Fax to list-of-numbers, where the second (and third?) will
>          only be used if the first number fails
> 
> No.

But it is a pretty good idea!

>     I hope to implement the client as a 'printer,' with a set of 
>     instructions in the first line(s), including recipient, phone
>     number (list), list of attachments, notification e-mail address, 
>     etc.  I don't really know how I will pull that off at this time, 
>     but that is my hope.

Dennis, may I suggest a CGI script and a web server? This seems tailor
made for a web page and forms to submit with the appropriate
fields. Take a look at ftp://ftp.sgi.com/sgi/fax/contrib/ and the
various contributed tools there.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNFdfQD/+ItycgIJRAQGA1wQAhYgekXcGIaYO1VfxEaY1fgpvmTF4pIa5
GJ1yinCncxaqpf/cSAbWrx6VwvJuAVVzU+Bdd2WaL0ztYspz10SZ6eimNiN2jk9H
SiLIXqVYuxaV0pKbgeawjdShz5N42l5nQprZ3BBB5WN7bKa7BjZXLwEXthgGYxlS
wKRrGsXarYQ=
=l2QY
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: "Roderic Simmons" <rod@designdata.com>
To: <flexfax@sgi.com>
Subject: flexfax: FAXQ
Date: Wed, 29 Oct 1997 09:21:07 -0500
Sender: owner-flexfax@celestial.com

Running Linux Redhat version 2.0.29

When I compose a message and send it to the fax it goes to the faxq in

 /var/spool/fax/sendq

the message never leaves the sendq

I compose the message in PcPine version 3.0 and send it to a person called
Test@server.domain.org

in my aliase file their is a statement :
Test: ":include:/home/staff/fax/ddsi.lst

in the /home/staff/fax
directory the file ddsi.lst states
"|/usr/local/bin/faxmail -d 'Test'@296-5381"

I do not know what to do

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Wed, 29 Oct 1997 11:13:20 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Ancil Bethelmy <cowpoke@wave.cape.net>
Cc: HylaFAX Mailing List <flexfax@sgi.com>
Subject: Re: flexfax: Need info on font installation
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Wed, 29 Oct 1997, Ancil Bethelmy wrote:

> I have completed the compilation and installation of most of the support
> files for setting up HylaFAX. I am now uncertain about what steps to
> take to install the fonts for ghostscript.
> 
> The list of font files include the 3.33, 4.0 and std fonts tar files. I
> have not seen any make files for doing the font file installation.

What operating system? The fonts go in one of the directories
specified in the Ghostscript installation as a fonts source: if you
type "gs -h", you should get a list of them along with all the active
drivers. Under SunOS, it's in /usr/local/share/ghostscript/fonts, for
example.

Also: I recommend the latest/greatest ghostscript, version 5.03,
available at ftp://ftp.cs.wisc.edu/pub/ghost/aladdin/current/. It
handles some printer drivers, etc. that are not in earlier
distributions.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNFdglT/+ItycgIJRAQGXngP/VSbwhrX0ADW3xDobXkqbBAk9kJgU9F0q
a/34m/ej61RJ0w1r7TgE+XQvwvFwB2AAH4Nkmv5+IUmJg7xwZzstplRKJZYEllN4
H4BZdPRRihc1VjpYe7TvPk02nswVg3mQ8/gx6WXsnXTwHVb3oASAmTKnNOyaP4hH
HuBFJyOQSNQ=
=c/rz
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Wed, 29 Oct 1997 12:14:50 -0500 (EST)
Organization: The Creative Group
From: Patrick Gardella <patrick@cre8tivegroup.com>
To: flexfax@sgi.com
Subject: flexfax: Doneq and Killed Faxes
Sender: owner-flexfax@celestial.com

I've got a local only email - fax gateway.  Works wonderfully, so no
problem there.

I am working on a perl script to handle all the faxes that can't be sent
for whatever reason, and print them to a laser printer for snail mail
delivery.  This script works well for "normal" errors (no answer, phase
B errors, etc).

But it does not work for faxes that are killed at the killtime.   They
are not moved to the doneq directory, as far as I can tell.

I emulate the faxqclean program (As best I can) to accomplish what I want.  I
search the doneq directory for a "status" line that is non-empty.  Any that
have a string get printed.  That, again, works just fine.

But is there anyway to know what faxes have been killed?  I get the error
messages by email (example below), but I want to automate this printing as well.

I appreciate the hints/help!

Patrick


Your facsimile job to xxx-xxx-xxxx could not be completed before the appointed
deadline.

    ---- Unsent job status ----

   Destination: xxx-xxx-xxxx
         JobID: 25465
       GroupID: 25465
        Sender: patrick
      Mailaddr: patrick@cre8tivegroup.com
Submitted From: localhost
    Page Width: 215 (mm)
   Page Length: 279 (mm)
    Resolution: 98 (lpi)
        Status:   (nothing available)
       Dialogs: 0 (exchanges with remote device)
         Dials: 0 (consecutive failed calls to destination)
         Calls: 0 (total phone calls placed)
         Pages: 0 (pages transmitted)
      TotPages: 0 (total pages to transmit)
      Attempts: 0 (attempts to send current page)
        Dirnum: 0 (directory of next page to send)

    ---- Documents submitted for transmission ----

The following documents were submitted for transmission and are
available on the server for reuse until they are automatically
purged when this job is removed.  Documents may also be manually
removed using the faxrm command; consult faxrm(1) for information.

Filename                 Size Type
docq/doc25443.ps.25465     8556 PostScript

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Doneq and Killed Faxes
To: patrick@cre8tivegroup.com (Patrick Gardella)
Date: Thu, 30 Oct 1997 08:46:11 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Patrick Gardella wrote:
    
    I've got a local only email - fax gateway.  Works wonderfully, so no
    problem there.
    
    I am working on a perl script to handle all the faxes that can't be sent
    for whatever reason, and print them to a laser printer for snail mail
    delivery.  This script works well for "normal" errors (no answer, phase
    B errors, etc).
    
    But it does not work for faxes that are killed at the killtime.   They
    are not moved to the doneq directory, as far as I can tell.

I've just run a test and the jobs are moved to the spool/doneq
if killtime arrived.
    
    I emulate the faxqclean program (As best I can) to accomplish what I want.  I
    search the doneq directory for a "status" line that is non-empty.  Any that
    have a string get printed.  That, again, works just fine.

The "status:" line in the spool/doneq/q#### is empty (I'll have a look
at the code if this should be changed).
    
    But is there anyway to know what faxes have been killed?  I get the error
    messages by email (example below), but I want to automate this printing as well.
    
    I appreciate the hints/help!

One thing that could be checked is the tts: and killtime: lines
and check if tts >= killtime (which isn't really waterproof);

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 30 Oct 1997 08:06:24 -0500 (EST)
Organization: The Creative Group
From: Patrick Gardella <patrick@cre8tivegroup.com>
To: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Doneq and Killed Faxes
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com


On 30-Oct-97 Matthias.Apitz@SOFTCON.de wrote:
>Patrick Gardella wrote:
>    
>    I've got a local only email - fax gateway.  Works wonderfully, so no
>    problem there.
>    
>    I am working on a perl script to handle all the faxes that can't be sent
>    for whatever reason, and print them to a laser printer for snail mail
>    delivery.  This script works well for "normal" errors (no answer, phase
>    B errors, etc).
>    
>    But it does not work for faxes that are killed at the killtime.   They
>    are not moved to the doneq directory, as far as I can tell.
>
>I've just run a test and the jobs are moved to the spool/doneq
>if killtime arrived.
>    
>    I emulate the faxqclean program (As best I can) to accomplish what I want. 
>  I search the doneq directory for a "status" line that is non-empty.  Any that
>    have a string get printed.  That, again, works just fine.
>
>The "status:" line in the spool/doneq/q#### is empty (I'll have a look
>at the code if this should be changed).

It would certainly help *me* if it should!  This is not one of the typical uses 
for the doneq, since many people will be sending from their other systems, and
could just resend when they get the email saying it didn't go.  

I've got killtime as +3 days, so it doesn't happen that often (only twice in
4 months).  

>    But is there anyway to know what faxes have been killed?  I get the error
>    messages by email (example below), but I want to automate this printing as 
well.
>    
>    I appreciate the hints/help!
>
>One thing that could be checked is the tts: and killtime: lines
>and check if tts >= killtime (which isn't really waterproof);

I'll implement it this way for now.  It's better to get two copies of the fax, 
instead of none.

Thanks for the help.

Patrick

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Tue, 28 Oct 1997 08:15:21 -0600 (CST)
From: "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
To: "Scott R. Keszler" <keszler@rrnet.com>
cc: flexfax@sgi.com
Subject: flexfax: Re: your mail
Sender: owner-flexfax@celestial.com

>From you log you can see that your modem sent a RTN, retrain
negative, asking the remote machine to send the page again (no clue why),
but the remote just hanged up. Hence hylafax terminates normally but no
page was received that was valid.

Julio

On Mon, 27 Oct 1997, Scott R. Keszler wrote:

> I'm new to this list and to hylaFAX.
> 
> Running hylafax-v4.0pl1 on Linux 1.3-3#17 on a Cyrix 6x86L (P166, not MMX)
> 
> hylaFAX configured, compiled, installed, and setup with no problems. It
> will be
> used primarily to receive.  Testing fax receive, I get a connection and fax
> negotiation succeeds.  I sent a one page test, following is the log from the
> beginning of the fax xmit:
> 
> >Oct 27 20:20:02.27: [  172]: RECV: begin page
> >Oct 27 20:20:02.27: [  172]: RECV: send trigger 022
> >Oct 27 20:20:26.38: [  172]: RECV: 28616 bytes of data, 1077 total lines
> >Oct 27 20:20:26.38: [  172]: --> [17:+FPTS: 2,2219,0,0]
> >Oct 27 20:20:27.61: [  172]: --> [7:+FET: 2]
> >Oct 27 20:20:27.61: [  172]: RECV recv EOP (no more pages or documents)
> >Oct 27 20:20:27.72: [  172]: --> [2:OK]
> >Oct 27 20:20:27.72: [  172]: RECV send RTN (retrain negative)
> >Oct 27 20:20:27.72: [  172]: <-- [7:AT+FDR\r]
> >Oct 27 20:20:30.44: [  172]: --> [8:+FHNG: 0]
> >Oct 27 20:20:30.44: [  172]: REMOTE HANGUP: Normal and proper end of
> connection (code 0)
> >Oct 27 20:20:30.44: [  172]: RECV FAX (00000021): recvq/fax00009.tif from
> 7012820410, route to <unspecified>, 0 pages in 0:34
> >Oct 27 20:20:30.44: [  172]: RECV FAX: Normal and proper end of connection
> >Oct 27 20:20:30.44: [  172]: RECV FAX (00000021): session with 7012820410
> terminated abnormally: Normal and proper end of connection
> >Oct 27 20:20:30.44: [  172]: RECV FAX: bin/faxrcvd "recvq/fax00009.tif"
> "ttyS2" "00000021" "Normal and proper end of connection"
> 
> Note that after +FHNG I get a code 0 "Normal and proper end of connection";
> then
> the session terminates ABnormally and apparantly throws away the .tif file.
>  By
> the time faxrcvd runs, the file no longer exists.
> 
> Why does "Normal and proper end of connection" then terminate abnormally?
> 
> 
> -- 
>           Scott R. Keszler
>           M.I.S. Manager
>           Lewis Transportation Systems
>           701-282-5330 x160
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
X-Authentication-Warning: tui.pinnacle.co.nz: jonc owned process doing -bs
Date: Wed, 29 Oct 1997 13:47:59 +1300 (NZDT)
From: Jonathan Chen <jonc@pinnacle.co.nz>
To: kevin@meso.com
cc: flexfax@sgi.com
Subject: Re: flexfax: Received Fax question 
Sender: owner-flexfax@celestial.com

On Fri, 24 Oct 1997 kevin@meso.com wrote:

> Hi folks,
>
> I'm using a MultiTech Systems MultiModem, MT2834ZPX.
> The config file was built using faxaddmodem without
> straying from the defaults.  However, when viewing incoming
> text faxes (which were sent as Postscript files) with
> Ghostscript (using either fax2ps or fax2tiff), the images
> looked distorted ("smudged" is an apropos term).  I edited
> the modem config file setting ModemRate down to 9600 and it
> looked much better.  I also tried leaving ModemRate at 38400
> and set ModemFlowControl to xonxoff instead of rtscts and it
> also looked fine.
>
> Is this an appropriate workaround?  Or have I created potential
> new problems?

Most modem chips only work with a max speed of 19200 with xon/xoff
flow control. However, it's always better to talk to modem at your
maximum possible speed with rts/cts flow control. To get around this
you can alter your config files to have the following:

    ModemType:          Class2  # use class 2 interface
    ModemRate:          38400   # locked for fax sending&receiving
    ModemFlowControl:   rtscts  # default
    #
    # Chip locks the line rate and switches to xon/xoff flow
    # control when receiving facsimile--we must follow suit.
    #
    ModemAnswerFaxBeginCmd: "<19200><xon>"  # switch line rate & flow control

Cheers.
--
Jonathan Chen <jonc@pinnacle.co.nz>                 Once is dumb luck.
                                                 Twice is coincidence.
             Three times and Somebody Is Trying To Tell You Something.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: "Wolfgang Baudler" <baudler@lulu.gsf.de>
Date: Thu, 30 Oct 1997 13:38:01 +0100
X-no-archive: yes
X-Face: '66dS68B{d&G{<va3%f=a`X5xCZI+=0(&BA8>FU5YVZ7[c,.Us2J*!JR`x~Ih=p`_o$j[byFh?+/zQ`dtF=|}=z!>q`-$k"SapvS?@G`.4%\]_25?yLyE]J31
Reply-To: baudler@gsf.de
To: flexfax@sgi.com
Subject: flexfax: sendpage leaves modem in state "Waiting for modem to come ready"
Cc: baudler@lulu.gsf.de
Sender: owner-flexfax@celestial.com

Hello,

after configuring HylaFAX for SMS paging I have the following problem:

I send a page for example with

sendpage -p Eplus12345 "Test"

The page gets delivered. See log below.
But then faxstat reports forever

HylaFAX scheduler on caravaggio.gsf.de: Running
Modem ttyf2 (+49.89.XXXXXXXX): Waiting for modem to come ready

Only a  "killall -HUP faxgetty" seems to reinitialize the modem
to the state "Running and idle".


Session Log:

Oct 30 13:26:29.50: [15486]: SESSION BEGIN 00001136 01771167
Oct 30 13:26:29.50: [15486]: SEND PAGE: JOB 314 DEST 001771167 COMMID 00001136
Oct 30 13:26:29.50: [15486]: MODEM set DTR OFF
Oct 30 13:26:29.50: [15486]: MODEM set baud rate: 0 baud (flow control
unchanged)
Oct 30 13:26:29.50: [15486]: DELAY 2600 ms
Oct 30 13:26:32.10: [15486]: MODEM set DTR ON
Oct 30 13:26:32.10: [15486]: MODEM set baud rate: 38400 baud, input flow
RTS/CTS, output flow RTS/CTS
Oct 30 13:26:32.11: [15486]: MODEM flush i/o
Oct 30 13:26:32.11: [15486]: <-- [39:AT&B1&N0&S0*F0S18=4S38.3=1E0V1Q0S0=0H0\r]
Oct 30 13:26:32.27: [15486]: --> [2:OK]
Oct 30 13:26:32.27: [15486]: <-- [21:ATS8=2S7=60&H3&D2&C1\r]
Oct 30 13:26:32.28: [15486]: --> [2:OK]
Oct 30 13:26:32.29: [15486]: <-- [5:ATM0\r]
Oct 30 13:26:32.29: [15486]: --> [2:OK]
Oct 30 13:26:32.29: [15486]: MODEM set parity: 8 bits, no parity
Oct 30 13:26:32.29: [15486]: MODEM input buffering enabled
Oct 30 13:26:32.30: [15486]: <-- [12:AT+FCLASS=0\r]
Oct 30 13:26:32.40: [15486]: --> [2:OK]
Oct 30 13:26:32.40: [15486]: <-- [19:AT&K0&N15B0E0&D2N0\r]
Oct 30 13:26:32.51: [15486]: --> [2:OK]
Oct 30 13:26:32.51: [15486]: DIAL 0W01771167
Oct 30 13:26:32.51: [15486]: <-- [15:ATDT0W01771167\r]
Oct 30 13:26:52.62: [15486]: --> [29:CONNECT 38400/V.22  1200/None]
Oct 30 13:26:52.62: [15486]: EXPECT ID (paging central identification)
Oct 30 13:26:52.62: [15486]: <-- <1:\015>
Oct 30 13:26:54.62: [15486]: MODEM TIMEOUT: reading line from modem
Oct 30 13:26:54.62: [15486]: <-- <1:\015>
Oct 30 13:26:56.62: [15486]: MODEM TIMEOUT: reading line from modem
Oct 30 13:26:56.62: [15486]: <-- <1:\015>
Oct 30 13:26:58.62: [15486]: MODEM TIMEOUT: reading line from modem
Oct 30 13:26:58.62: [15486]: <-- <1:\015>
Oct 30 13:27:00.62: [15486]: MODEM TIMEOUT: reading line from modem
Oct 30 13:27:00.62: [15486]: <-- <1:\015>
Oct 30 13:27:02.62: [15486]: MODEM TIMEOUT: reading line from modem
Oct 30 13:27:02.62: [15486]: --> <3:ID=>
Oct 30 13:27:02.62: [15486]: RECV ID ("ID=")
Oct 30 13:27:02.62: [15486]: SEND device identification/login request
Oct 30 13:27:02.62: [15486]: <-- <5:\033PG1\015>
Oct 30 13:27:03.30: [15486]: --> <1:\006>
Oct 30 13:27:03.30: [15486]: RECV ACK (login successful)
Oct 30 13:27:03.40: [15486]: --> <3:\033[p>
Oct 30 13:27:03.40: [15486]: RECV go-ahead (prologue done)
Oct 30 13:27:03.40: [15486]: SEND message block
Oct 30 13:27:03.40: [15486]: <-- <22:\0024917712345\015Test\015\0033<:\015>
Oct 30 13:27:05.48: [15486]: --> <68:Message 916661303 send successful -
message submitted for processing>
Oct 30 13:27:05.48: [15486]: RECV: Message 916661303 send successful - message
submitted for processing
Oct 30 13:27:05.48: [15486]: --> <1:\006>
Oct 30 13:27:05.48: [15486]: RECV ACK (message block accepted)
Oct 30 13:27:05.48: [15486]: <-- <2:\004\015>
Oct 30 13:27:06.14: [15486]: --> <2:\033\004>
Oct 30 13:27:06.14: [15486]: RECV EOT (disconnect)
Oct 30 13:27:06.14: [15486]: SEND PAGE: FROM baudler@lulu.gsf.de TO 001771167
(sent in 0:34)
Oct 30 13:27:06.14: [15486]: <-- [5:ATH0\r]
Oct 30 13:27:11.14: [15486]: MODEM TIMEOUT: reading line from modem
Oct 30 13:27:11.14: [15486]: MODEM <Timeout>
Oct 30 13:27:11.14: [15486]: MODEM input buffering enabled
Oct 30 13:27:11.14: [15486]: MODEM set DTR OFF
Oct 30 13:27:11.14: [15486]: MODEM set baud rate: 0 baud (flow control
unchanged)
Oct 30 13:27:11.14: [15486]: SESSION END


How can I configure the paging system so that it sets the modem back to
normal state like the fax system does?

Any help is welcome...


PS: System is HylaFAX v4.0pl1, running on an SGI O2 with IRIX 6.3

Wolfgang





-- 
----------------------------------------------------------------------
Wolfgang Baudler
mailto:baudler@gsf.de  Tel.: +49.89.31874463
                       Fax.: +49.89.31874243
Home-Page: http://lulu.gsf.de:8888        
----------------------------------------------------------------------
.. Our continuing mission: To seek out knowledge of C, to explore
strange UNIX commands, and to boldly code where no one has man page 4.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: sendpage leaves modem in state "Waiting for modem to come ready"
To: baudler@gsf.de
Date: Thu, 30 Oct 1997 18:00:37 +0100 (MEZ)
Cc: flexfax@sgi.com, baudler@lulu.gsf.de
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Wolfgang Baudler wrote:
    
    Hello,
    
    after configuring HylaFAX for SMS paging I have the following problem:
    
    I send a page for example with
    
    sendpage -p Eplus12345 "Test"
    
    The page gets delivered. See log below.
    But then faxstat reports forever
    
    HylaFAX scheduler on caravaggio.gsf.de: Running
    Modem ttyf2 (+49.89.XXXXXXXX): Waiting for modem to come ready
    
    Only a  "killall -HUP faxgetty" seems to reinitialize the modem
    to the state "Running and idle".
    
    
    Session Log:
    
    <... log deleted ...>

BTW: You should set your ModemDialCmd: to "ATDT0W%s" to
avoid setting the "0W" into pagermap(4F) file (just a hint and
has nothing to do with your problem).

    Oct 30 13:26:32.40: [15486]: <-- [19:AT&K0&N15B0E0&D2N0\r]
    Oct 30 13:26:32.51: [15486]: --> [2:OK]
    Oct 30 13:26:32.51: [15486]: DIAL 0W01771167
    Oct 30 13:26:32.51: [15486]: <-- [15:ATDT0W01771167\r]
    
    <... log deleted ...>
    
    How can I configure the paging system so that it sets the modem back to
    normal state like the fax system does?
    
    Any help is welcome...

Please enable and show a serverTracing for faxgetty(1M).

	matthias
    
    PS: System is HylaFAX v4.0pl1, running on an SGI O2 with IRIX 6.3
    
    Wolfgang
    
    
    
    
    
    -- 
    ----------------------------------------------------------------------
    Wolfgang Baudler
    mailto:baudler@gsf.de  Tel.: +49.89.31874463
                           Fax.: +49.89.31874243
    Home-Page: http://lulu.gsf.de:8888        
    ----------------------------------------------------------------------
    .. Our continuing mission: To seek out knowledge of C, to explore
    strange UNIX commands, and to boldly code where no one has man page 4.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: "Wolfgang Baudler" <baudler@lulu.gsf.de>
Date: Fri, 31 Oct 1997 09:19:20 +0100
X-no-archive: yes
X-Face: '66dS68B{d&G{<va3%f=a`X5xCZI+=0(&BA8>FU5YVZ7[c,.Us2J*!JR`x~Ih=p`_o$j[byFh?+/zQ`dtF=|}=z!>q`-$k"SapvS?@G`.4%\]_25?yLyE]J31
Reply-To: baudler@gsf.de
To: Matthias.Apitz@SOFTCON.de (Matthias Apitz), flexfax@sgi.com
Subject: Re: flexfax: sendpage leaves modem in state "Waiting for modem to come ready"
Sender: owner-flexfax@celestial.com

On Oct 30,  6:00pm, Matthias.Apitz@SOFTCON.de wrote:
> Subject: Re: flexfax: sendpage leaves modem in state "Waiting for modem to
> Wolfgang Baudler wrote:
>
>     Hello,
>
>     after configuring HylaFAX for SMS paging I have the following problem:
>
>     I send a page for example with
>
>     sendpage -p Eplus12345 "Test"
>
>     The page gets delivered. See log below.
>     But then faxstat reports forever
>
>     HylaFAX scheduler on caravaggio.gsf.de: Running
>     Modem ttyf2 (+49.89.XXXXXXXX): Waiting for modem to come ready
>
>     Only a  "killall -HUP faxgetty" seems to reinitialize the modem
>     to the state "Running and idle".
>
>     How can I configure the paging system so that it sets the modem back to
>     normal state like the fax system does?
>
>     Any help is welcome...
>
> Please enable and show a serverTracing for faxgetty(1M).
>

OK, here is the serverTracing from initializing the system until one
paging job is sent and nothing more happens...
Sorry if this is a bit long...

Oct 31 09:06:43 6F:caravaggio syslogd: restart
Oct 31 09:06:54 3D:caravaggio FaxQueuer[17629]: servertracing = 65535 (line 7)
Oct 31 09:06:54 3D:caravaggio FaxQueuer[17629]: maxdials = 1 (line 8)
Oct 31 09:06:54 6D:caravaggio FaxQueuer[17629]: HylaFAX (tm) Version 4.0pl1
Oct 31 09:06:54 6D:caravaggio FaxQueuer[17629]: Copyright (c) 1990-1996 Sam
Leffler
Oct 31 09:06:54 6D:caravaggio FaxQueuer[17629]: Copyright (c) 1991-1996 Silicon
Graphics, Inc.
Oct 31 09:06:54 6V:caravaggio FaxGetty[17598]: HELLO
Oct 31 09:06:54 6D:caravaggio FaxQueuer[17629]: FIFO RECV
"+ttyf2:N+49.89.31874243"
Oct 31 09:06:54 6D:caravaggio FaxQueuer[17629]: MODEM ttyf2: NUMBER
+49.89.31874243
Oct 31 09:06:54 6D:caravaggio FaxQueuer[17629]: FIFO RECV
"+ttyf2:RP3fd7e7ff:ff"
Oct 31 09:06:54 6D:caravaggio FaxQueuer[17629]: MODEM ttyf2: READY,
capabilities P3fd7e7ff:ff
Oct 31 09:06:54 6D:caravaggio FaxGetty[17633]: servertracing = 65535 (line 47)
Oct 31 09:06:54 6D:caravaggio FaxGetty[17633]: sessiontracing = 4111 (line 48)
Oct 31 09:06:54 6D:caravaggio FaxGetty[17633]: recvfilemode = 0644 (line 49)
Oct 31 09:06:54 6D:caravaggio FaxGetty[17633]: logfilemode = 0644 (line 50)
Oct 31 09:06:54 6D:caravaggio FaxGetty[17633]: devicemode = 0644 (line 51)
Oct 31 09:06:54 6D:caravaggio FaxGetty[17633]: ringsbeforeanswer = 1 (line 52)
Oct 31 09:06:54 6D:caravaggio FaxGetty[17633]: speakervolume = off (line 53)
Oct 31 09:06:54 6D:caravaggio FaxGetty[17633]: gettyargs = -h %l dx_%s (line
54)
Oct 31 09:06:54 6D:caravaggio FaxGetty[17633]: localidentifier =
+49.89.XXXXXXXX (line 55)
Oct 31 09:06:54 6D:caravaggio FaxGetty[17633]: taglinefont = etc/lutRS18.pcf
(line 56)
Oct 31 09:06:54 6D:caravaggio FaxGetty[17633]: taglineformat = From %%l (GSF
Medis M3)|%c|Page %%p of %%t (line 57)
Oct 31 09:06:54 6D:caravaggio FaxGetty[17633]: nocarrierretrys = 1 (line 58)
Oct 31 09:06:54 6D:caravaggio FaxGetty[17633]: maxrecvpages = 25 (line 59)
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: logfacility = local5 (line 60)
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: modemtype = Class2.0 (line 66)
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: modemrate = 38400 (line 67)
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: modemflowcontrol = rtscts (line
68)
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: modemnoflowcmd = AT&H0 (line 70)
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: modemhardflowcmd = AT&H3 (line
71)
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: modemsoftflowcmd = AT&H4 (line
72)
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: modemsetupdtrcmd = AT&D2 (line
73)
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: modemsetupdcdcmd = AT&C1 (line
74)
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: modemsetupaacmd = AT+FAA=1 (line
76)
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: modemresetcmds =
AT&B1&N0&S0*F0S18=4S38.3=1 (line 88)
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: modemdialcmd = ATDT%s (line 90)
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: nocarrierretrys = 3 (line 91)
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: modemsetvolumecmd = ATM0 ATM1L1
ATM1L3 ATM1L5 ATM1L7 (line 93)
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: polllockwait = 10 (line 95)
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: pollmodemwait = 10 (line 96)
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: class2apcmd =  (line 98)
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: class2apquerycmd =  (line 99)
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: Unknown configuration parameter
"pagersetupcmds" ignored at line 108
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: Unknown configuration parameter
"pagermaxmsglength" ignored at line 109
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: Unknown configuration parameter
"pagerttyparity" ignored at line 110
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: Unknown configuration parameter
"ixoidtimeout" ignored at line 116
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: Unknown configuration parameter
"ixologintimeout" ignored at line 118
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: Unknown configuration parameter
"ixogatimeout" ignored at line 119
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: Unknown configuration parameter
"ixoxmittimeout" ignored at line 121
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: Unknown configuration parameter
"ixoacktimeout" ignored at line 122
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: OPEN /dev/ttyf2
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: MODEM set DTR OFF
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: MODEM set baud rate: 0 baud
(flow control unchanged)
Oct 31 09:06:54 6D:caravaggio FaxQueuer[17629]: FIFO RECV
"+ttyf2:N+49.89.31874243"
Oct 31 09:06:54 6D:caravaggio FaxQueuer[17629]: MODEM ttyf2: NUMBER
+49.89.31874243
Oct 31 09:06:54 6V:caravaggio FaxGetty[17633]: DELAY 2600 ms
Oct 31 09:06:55 5D:caravaggio HylaFAX[17631]: HylaFAX INET Protocol Server:
restarted.
Oct 31 09:06:55 5D:caravaggio HylaFAX[17631]: HylaFAX Old Protocol Server:
restarted.
Oct 31 09:06:55 5D:caravaggio HylaFAX[17631]: HylaFAX SNPP Protocol Server:
restarted.
Oct 31 09:06:57 6V:caravaggio FaxGetty[17633]: MODEM set DTR ON
Oct 31 09:06:57 6V:caravaggio FaxGetty[17633]: MODEM set baud rate: 38400 baud,
input flow RTS/CTS, output flow RTS/CTS
Oct 31 09:06:57 6V:caravaggio FaxGetty[17633]: MODEM flush i/o
Oct 31 09:06:57 6V:caravaggio FaxGetty[17633]: <--
[39:AT&B1&N0&S0*F0S18=4S38.3=1E0V1Q0S0=0H0\r]
Oct 31 09:06:57 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:06:57 6V:caravaggio FaxGetty[17633]: <-- [21:ATS8=2S7=60&H3&D2&C1\r]
Oct 31 09:06:57 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: <-- [12:AT+FCLASS=?\r]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: --> [13:0,2,2.0,6,8,Z]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports "Data"
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports "Class 2"
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports "Class 2.0"
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports "Voice"
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: <-- [14:AT+FCLASS=2.0\r]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: <-- [8:AT+FMI?\r]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: --> [5:ZyXEL]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM: Mfr ZyXEL
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: <-- [8:AT+FMM?\r]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: --> [18:E2864  V 1.18     ]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: <-- [8:AT+FMR?\r]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: --> [18:E2864  V 1.18     ]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM: Model E2864
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM: Revision V 1.18
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: <-- [9:AT+FCC=?\r]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: -->
[43:(0,1),(0-5),(0-2),(0-2),(0,1),(0),(0),(0-7)]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports 3.85 line/mm
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports 7.7 line/mm
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports 2400 bit/s
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports 4800 bit/s
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports 7200 bit/s
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports 9600 bit/s
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports 12000 bit/s
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports 14400 bit/s
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports page width 1728
pixels in 215 mm
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports page width 2048
pixels in 255 mm
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports page width 2432
pixels in 303 mm
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports A4 page length
(297 mm)
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports B4 page length
(364 mm)
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports unlimited page
length
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports 1-D MR
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports 2-D MR
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports 0 ms/scanline
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports 5 ms/scanline
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports 10 ms, 5
ms/scanline
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports 10 ms/scanline
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports 20 ms, 10
ms/scanline
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports 20 ms/scanline
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports 40 ms, 20
ms/scanline
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports 40 ms/scanline
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: <-- [9:AT+FCQ=?\r]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: --> [11:(0-2),(0-2)]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports 1D+2D copy
quality checking (but not enabled)
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports subaddressing
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: MODEM Supports selective polling
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: <-- [9:AT+FSP=?\r]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: --> [3:0,1]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:06:58 6V:caravaggio FaxGetty[17633]: <-- [14:AT+FCLASS=2.0\r]
Oct 31 09:06:59 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:06:59 6V:caravaggio FaxGetty[17633]: <-- [9:AT+FLO=2\r]
Oct 31 09:06:59 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:06:59 6V:caravaggio FaxGetty[17633]: <-- [9:AT+FPP=0\r]
Oct 31 09:06:59 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:06:59 6V:caravaggio FaxGetty[17633]: <-- [9:AT+FBO=0\r]
Oct 31 09:06:59 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:06:59 6V:caravaggio FaxGetty[17633]: <-- [10:AT+FCT=30\r]
Oct 31 09:06:59 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:06:59 6V:caravaggio FaxGetty[17633]: <-- [15:AT+FNR=1,1,1,1\r]
Oct 31 09:06:59 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:06:59 6V:caravaggio FaxGetty[17633]: <-- [9:AT+FIE=0\r]
Oct 31 09:06:59 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:06:59 6V:caravaggio FaxGetty[17633]: <--
[23:AT+FCC=1,5,2,2,1,0,0,0\r]
Oct 31 09:06:59 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:06:59 6V:caravaggio FaxGetty[17633]: MODEM ZYXEL E2864/V 1.18
Oct 31 09:06:59 6V:caravaggio FaxGetty[17633]: <-- [5:ATM0\r]
Oct 31 09:07:00 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:07:00 6V:caravaggio FaxGetty[17633]: <-- [9:AT+FCR=1\r]
Oct 31 09:07:00 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:07:00 6V:caravaggio FaxGetty[17633]: <--
[25:AT+FLI="+49.89.31874243"\r]
Oct 31 09:07:00 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:07:00 6V:caravaggio FaxGetty[17633]: <-- [9:AT+FAA=1\r]
Oct 31 09:07:00 6V:caravaggio FaxGetty[17633]: --> [2:OK]
Oct 31 09:07:00 6V:caravaggio FaxGetty[17633]: STATE CHANGE: BASE -> RUNNING
Oct 31 09:07:00 6V:caravaggio FaxGetty[17633]: MODEM input buffering disabled
Oct 31 09:07:00 6D:caravaggio FaxQueuer[17629]: FIFO RECV
"+ttyf2:RP3fd7e7ff:ff"
Oct 31 09:07:00 6D:caravaggio FaxQueuer[17629]: MODEM ttyf2: READY,
capabilities P3fd7e7ff:ff

Init complete... Here begins the paging job:


Oct 31 09:07:09 6D:caravaggio HylaFAX[17639]: Filesystem has SysV-style file
creation semantics.
Oct 31 09:07:09 6D:caravaggio FaxQueuer[17629]: FIFO RECV "Tclient/17639:J0010"
Oct 31 09:07:09 6D:caravaggio FaxQueuer[17629]: TRIGGER J0010
Oct 31 09:07:09 6D:caravaggio FaxQueuer[17629]: FIFO RECV "Sclient/17639:316"
Oct 31 09:07:09 6D:caravaggio FaxQueuer[17629]: SUBMIT JOB 316
Oct 31 09:07:09 6D:caravaggio FaxQueuer[17629]: JOB 316 (suspended dest  pri
127 tts 0:00 killtime 5:00): CREATE
Oct 31 09:07:09 6D:caravaggio FaxQueuer[17629]: JOB 316 (ready dest +01771167
pri 127 tts 0:00 killtime 5:00): READY
Oct 31 09:07:09 6D:caravaggio FaxQueuer[17629]: FIFO SEND client/17639 msg "S*"
Oct 31 09:07:09 6D:caravaggio FaxQueuer[17629]: JOB 316 (ready dest +01771167
pri 127 tts 0:00 killtime 5:00): PROCESS
Oct 31 09:07:09 6D:caravaggio FaxQueuer[17629]: JOB 316 (active dest +01771167
pri 127 tts 0:00 killtime 5:00): ACTIVE
Oct 31 09:07:10 6D:caravaggio FaxQueuer[17629]: JOB 316 (active dest +01771167
pri 127 tts 0:00 killtime 4:59): CMD START /usr/local/sbin/pagesend -m ttyf2
sendq/q316 (PID 17640)
Oct 31 09:07:10 6D:caravaggio PageSend[17640]: servertracing = 65535 (line 47)
Oct 31 09:07:10 6D:caravaggio PageSend[17640]: sessiontracing = 4111 (line 48)
Oct 31 09:07:10 6D:caravaggio PageSend[17640]: recvfilemode = 0644 (line 49)
Oct 31 09:07:10 6D:caravaggio PageSend[17640]: logfilemode = 0644 (line 50)
Oct 31 09:07:10 6D:caravaggio PageSend[17640]: devicemode = 0644 (line 51)
Oct 31 09:07:10 6D:caravaggio PageSend[17640]: Unknown configuration parameter
"ringsbeforeanswer" ignored at line 52
Oct 31 09:07:10 6D:caravaggio PageSend[17640]: speakervolume = off (line 53)
Oct 31 09:07:10 6D:caravaggio PageSend[17640]: Unknown configuration parameter
"gettyargs" ignored at line 54
Oct 31 09:07:10 6D:caravaggio PageSend[17640]: localidentifier =
+49.89.XXXXXXXX (line 55)
Oct 31 09:07:10 6D:caravaggio PageSend[17640]: taglinefont = etc/lutRS18.pcf
(line 56)
Oct 31 09:07:10 6D:caravaggio PageSend[17640]: taglineformat = From %%l (GSF
Medis M3)|%c|Page %%p of %%t (line 57)
Oct 31 09:07:10 6D:caravaggio PageSend[17640]: nocarrierretrys = 1 (line 58)
Oct 31 09:07:10 6D:caravaggio PageSend[17640]: maxrecvpages = 25 (line 59)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: logfacility = local5 (line 60)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: modemtype = Class2.0 (line 66)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: modemrate = 38400 (line 67)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: modemflowcontrol = rtscts (line
68)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: modemnoflowcmd = AT&H0 (line 70)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: modemhardflowcmd = AT&H3 (line
71)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: modemsoftflowcmd = AT&H4 (line
72)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: modemsetupdtrcmd = AT&D2 (line
73)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: modemsetupdcdcmd = AT&C1 (line
74)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: modemsetupaacmd = AT+FAA=1 (line
76)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: modemresetcmds =
AT&B1&N0&S0*F0S18=4S38.3=1 (line 88)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: modemdialcmd = ATDT%s (line 90)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: nocarrierretrys = 3 (line 91)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: modemsetvolumecmd = ATM0 ATM1L1
ATM1L3 ATM1L5 ATM1L7 (line 93)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: polllockwait = 10 (line 95)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: pollmodemwait = 10 (line 96)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: class2apcmd =  (line 98)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: class2apquerycmd =  (line 99)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: pagersetupcmds =
AT&K0&N15B0E0&D2N0 (line 108)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: pagermaxmsglength = 128 (line
109)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: pagerttyparity = none (line 110)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: ixoidtimeout = 60 (line 116)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: ixologintimeout = 60 (line 118)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: ixogatimeout = 60 (line 119)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: ixoxmittimeout = 60 (line 121)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: ixoacktimeout = 60 (line 122)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: MODEM set DTR OFF
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: MODEM set baud rate: 0 baud
(flow control unchanged)
Oct 31 09:07:10 6V:caravaggio PageSend[17640]: DELAY 2600 ms
Oct 31 09:07:12 6V:caravaggio PageSend[17640]: MODEM set DTR ON
Oct 31 09:07:12 6V:caravaggio PageSend[17640]: MODEM set baud rate: 38400 baud,
input flow RTS/CTS, output flow RTS/CTS
Oct 31 09:07:12 6V:caravaggio PageSend[17640]: MODEM flush i/o
Oct 31 09:07:12 6V:caravaggio PageSend[17640]: <--
[39:AT&B1&N0&S0*F0S18=4S38.3=1E0V1Q0S0=0H0\r]
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: --> [2:OK]
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: <-- [21:ATS8=2S7=60&H3&D2&C1\r]
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: --> [2:OK]
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: <-- [12:AT+FCLASS=?\r]
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: --> [13:0,2,2.0,6,8,Z]
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: --> [2:OK]
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: MODEM Supports "Data"
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: MODEM Supports "Class 2"
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: MODEM Supports "Class 2.0"
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: MODEM Supports
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: MODEM Supports "Voice"
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: <-- [12:AT+FCLASS=0\r]
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: --> [2:OK]
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: <-- [6:AT&H3\r]
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: --> [2:OK]
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: MODEM: Mfr
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: MODEM  /
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: <-- [5:ATM0\r]
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: --> [2:OK]
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: MODEM set parity: 8 bits, no
parity
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: STATE CHANGE: BASE -> RUNNING
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: MODEM input buffering disabled
Oct 31 09:07:13 6V:caravaggio PageSend[17640]: SEND PAGE: JOB 316 DEST
001771167 COMMID 00001138
Oct 31 09:07:49 6V:caravaggio PageSend[17640]: SEND PAGE: FROM
baudler@lulu.gsf.de TO 001771167 (sent in 0:33)
Oct 31 09:07:54 6D:caravaggio FaxQueuer[17629]: JOB 316 (active dest +01771167
pri 127 tts 0:00 killtime 4:15): CMD DONE: exit status 0x200
Oct 31 09:07:54 6D:caravaggio FaxQueuer[17629]: JOB 316 (done dest +01771167
pri 127 tts 0:00 killtime 4:15): SEND DONE: 0:45
Oct 31 09:07:54 6D:caravaggio FaxQueuer[17629]: JOB 316 (done dest +01771167
pri 127 tts 0:00 killtime 4:15): DEAD
Oct 31 09:07:54 6D:caravaggio FaxQueuer[17629]: JOB 316 (done dest +01771167
pri 127 tts 0:00 killtime 4:15): DELETE
Oct 31 09:07:54 6D:caravaggio FaxQueuer[17629]: FIFO RECV "D0"
Oct 31 09:07:54 6D:caravaggio FaxQueuer[17629]: DELETE 0


Nothing happens after this point...


Wolfgang


-- 
----------------------------------------------------------------------
Wolfgang Baudler
mailto:baudler@gsf.de  Tel.: +49.89.31874463
                       Fax.: +49.89.31874243
Home-Page: http://lulu.gsf.de:8888        
----------------------------------------------------------------------
.. Our continuing mission: To seek out knowledge of C, to explore
strange UNIX commands, and to boldly code where no one has man page 4.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: sendpage leaves modem in state "Waiting for modem to come ready"
To: baudler@gsf.de
Date: Fri, 31 Oct 1997 18:21:08 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Wolfgang Baudler wrote:
    
    On Oct 30,  6:00pm, Matthias.Apitz@SOFTCON.de wrote:
    > Subject: Re: flexfax: sendpage leaves modem in state "Waiting for modem to
    > Wolfgang Baudler wrote:
    >
    >     Hello,
    >
    >     after configuring HylaFAX for SMS paging I have the following problem:
    >
    >     I send a page for example with
    >
    >     sendpage -p Eplus12345 "Test"
    >
    >     The page gets delivered. See log below.
    >     But then faxstat reports forever
    >
    >     HylaFAX scheduler on caravaggio.gsf.de: Running
    >     Modem ttyf2 (+49.89.XXXXXXXX): Waiting for modem to come ready
    >
    >     Only a  "killall -HUP faxgetty" seems to reinitialize the modem
    >     to the state "Running and idle".
    >
    >     How can I configure the paging system so that it sets the modem back to
    >     normal state like the fax system does?
    >
    >     Any help is welcome...
    >
    > Please enable and show a serverTracing for faxgetty(1M).
    >
    
    OK, here is the serverTracing from initializing the system until one
    paging job is sent and nothing more happens...
    Sorry if this is a bit long...
    
    <... trace deleted ...>

I've checked it and there isn't anything unusual; pagesend(1M)
should remove the modem lock-file after sending the message to the paging
terminal and faxgetty(1M) should monitor the same lock-file
and do validation of the pid in the lock-file every 30 seconds;

please check after a run of pagesend(1M) if the lock-file still
exists and what faxgetty(1M) is doing (use truss or strace
if IRIX comes with it to monitor the syscalls of the proc);

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 30 Oct 1997 12:09:57 -0300 (GMT+3)
From: Ben Parker <ben@dha.unon.org>
To: Lou Katz <lou@metron.com>
cc: flexfax@sgi.com
Subject: Re: flexfax: Setting outbound defaults on fax->mail
Sender: owner-flexfax@celestial.com

Depending on how you are doing the gateway, you need to create these
switches on the command lines of both faxmail and sendfax. Faxmail
controls the font (-f and -p), while sendfax controls resolution (-m =
fine), and mail notification (-DR). The -DR notifies both a re-queue as
well as success or failure...

/usr/local/bin/faxmail -n -f Courier-Bold -p 12 |
/usr/local/bin/sendfax -DR -m -d 'joe@5551212' (probably other switches
here like -f 'From Address <from@from.com>')


Ben

On Wed, 29 Oct 1997, Lou Katz wrote:

> I am running hylafax-v4.0pl1 on BSD/OS2.1 with sendmail-8.8.7.
> The mail-to-fax gateway works fine (i.e. mail joe@5551212.fax causes a
> fax to be sent to 5551212 addressed to joe).
> 
> I want three things to happen by default, but can't figure out where
> to put the flags, etc.
> 
> 1. Use fine mode.
> 2. Use a particular body font for the message (the courier comes out
>    too light).
> 3. Mail notify the sender on success as well as failure.
> 
> Hints or pointers to the proper man pages appreciated. I have read
> the docs many times, and if the info is there, I didn't recognize it.
> 
> 
> -=[Lou Katz]=-
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 31 Oct 1997 14:55:15 +0100
From: "Ferenc Cserei" <feri@graphisoft.hu>
To: flexfax@sgi.com
Subject: flexfax: problem sending faxes to pxb
X-Authenticated: fcserei by imi.graphisoft.hu
X-Licensed-To: Ferenc Cserei
Sender: owner-flexfax@celestial.com

We encuntered a problem when our properly running hylafax tries to send
faxses to places where small (home) pbx installed. 
It seems the pbx expects some kind of signalling from the sender to decide
wether it's a voice, data or fax call. Hylafax is quiet and the receiver
hangs the line, resulting no answer from remote.

Is there a way to configure hylafax to send this signal (a kind of repeating
beep) after the receiver picks up the phone.

Thanks

Ferenc Cserei
NIS engineer
feri@graphisoft.hu

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: problem sending faxes to pxb
To: feri@graphisoft.hu (Ferenc Cserei)
Date: Sat, 1 Nov 1997 07:19:58 +0000 (GMT)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> 
> Is there a way to configure hylafax to send this signal (a kind of repeating
> beep) after the receiver picks up the phone.

Hylafax only indirectly controls this. First check with you modem manual as
to whether it is possible to make the modem send fax calling tone.  If it
is, and it is the default, find the part of the modem initialisation string
which turns it off.  If it is capable, but off by default, add the relevant
sequence to the modem initialisation string.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: problem sending faxes to pxb
To: feri@graphisoft.hu (Ferenc Cserei)
Date: Sat, 1 Nov 1997 10:14:31 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Ferenc Cserei wrote:
    
    We encuntered a problem when our properly running hylafax tries to send
    faxses to places where small (home) pbx installed. 
    It seems the pbx expects some kind of signalling from the sender to decide
    wether it's a voice, data or fax call. Hylafax is quiet and the receiver
    hangs the line, resulting no answer from remote.
    
    Is there a way to configure hylafax to send this signal (a kind of         repeating beep) after the receiver picks up the phone.

This tone is called the CNG (1100 Hz, remains on for 1/2 s and goes
off for 3 s). It's not issued by HylaFAX itself. faxsend(1M) only dials
the number and your (fax-) modem should emit this tone. Check
your modem manual how to configure your modem to emitt this tone
and place the AT-command into the modem config-file (e.g. into
ModemResetCmds:)

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Steve Phillips <steve@bcs.co.za>
X-Openmail-Hops: 1
Date: Sat, 1 Nov 97 11:36:13 +0200
Subject: RE: flexfax: problem sending faxes to pxb
To: feri@graphisoft.hu, flexfax@sgi.com
Sender: owner-flexfax@celestial.com


Ferenc

It seems the pbx expects some kind of signalling from the sender to 
decidewether it's a voice, data or fax call. Hylafax is quiet and the 
receiver
hangs the line, resulting no answer from remote.

In your modem config file take out the @ in the ModemDialCmd :

ModemDialCmd:           ATDT%s          # T for tone dialing, taken @ 
out


Hi,

I re-check everything, the ~~~/sbin/hfaxd is the only file that matter in this case.

Nico Garcia wrote:

> On Fri, 31 Oct 1997, Uri Shkolnik wrote:
>
> > Hi Alan.
> >
> > I just tried to run the HylaFAX with your patch, but it seems that not only it not fix the wfhc bug, it also disabled hfaxd
> > from response to other clients (faxstat for example).
>
> Make sure you didn't overwrite your HylaFAX "hosts" file....
>
>                         Nico Garcia
>                         Engineer, CIRL
>                         Mass. Eye and Ear Infirmary
>                         raoul@cirl.meei.harvard.edu

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Comments: Authenticated sender is <spine001@gold.tc.umn.edu>
From: Julio C Spinelli <spine001@gold.tc.umn.edu>
To: Matthias.Apitz@SOFTCON.de
Date: Sat, 1 Nov 1997 10:56:38 +0000
Subject: Re: flexfax: problem sending faxes to pxb
CC: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

To have the modem emit this tone, you need to have autoanswer off, 
something like AT+FAA=0 works in many modems. To have this be part of 
the hylafax initialization, you need to add a statement similar 
(please check the manual pages for setup for the exact spelling) to

AdaptiveAnswer: AT+FAA=0 #turns adaptive answer off

To your config.xxx file (xxx represents the device name where the 
modem is) in the /var/spool/fax/etc directory.

Do this and it will work in many home pbxs. There are still some, 
that for reasons unknown to me confuse the modem into getting a busy 
tone and it just hangs up when they answer. Does anybody have any 
solutions for this?

Good luck

Julio
> From:          Matthias.Apitz@SOFTCON.de
> Subject:       Re: flexfax: problem sending faxes to pxb
> To:            feri@graphisoft.hu (Ferenc Cserei)
> Date:          Sat, 1 Nov 1997 10:14:31 +0100 (MEZ)
> Cc:            flexfax@sgi.com
> Reply-to:      Matthias.Apitz@SOFTCON.de (Matthias Apitz)

> Ferenc Cserei wrote:
>     
>     We encuntered a problem when our properly running hylafax tries to send
>     faxses to places where small (home) pbx installed. 
>     It seems the pbx expects some kind of signalling from the sender to decide
>     wether it's a voice, data or fax call. Hylafax is quiet and the receiver
>     hangs the line, resulting no answer from remote.
>     
>     Is there a way to configure hylafax to send this signal (a kind of repeating
>     beep) after the receiver picks up the phone.
> 
> This tone is called the CNG (1100 Hz, remains on for 1/2 s and goes
> off for 3 s). It's not issued by HylaFAX itself. faxsend(1M) only dials
> the number and your (fax-) modem should emitt this tone. Check
> your modem manual how to configure your modem to emitt this tone
> and place the AT-command into the modem config-file (e.g. into
> ModemResetCmds:)
> 
> 	matthias
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: problem sending faxes to pxb
To: spine001@gold.tc.umn.edu (Julio C Spinelli)
Date: Sat, 1 Nov 1997 18:21:35 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Julio C Spinelli wrote:
    
    To have the modem emit this tone, you need to have autoanswer off, 
    something like AT+FAA=0 works in many modems. To have this be part of 
    the hylafax initialization, you need to add a statement similar 
    (please check the manual pages for setup for the exact spelling) to
    
    AdaptiveAnswer: AT+FAA=0 #turns adaptive answer off
    
    To your config.xxx file (xxx represents the device name where the 
    modem is) in the /var/spool/fax/etc directory.

AdaptiveAnswer: controls as a boolean value whether or not an
adaptive answer strategy is used (and defaults to No). It has
nothing to do with the CNG tone and setting AT+FAA=0 into
this configuration parameter would not work. The problem of
the original question was also how to make the modem emit CNG
during a call. AT+FAA=1 is typically set in ModemSetupAACmd:
to make HylaFAX service a fax, data, or voice call
as identified by the modem(!). Naturally this will only work if the
supports a good adaptive-answering facility (for more infos
see html/setup-advanced.html and the man page config(4F)).

I don't know if or if not AT+FAA=0 in ModemSetupAACmd: for the
modem (NB: the original poster didn't give much information
about the modem) will switch on CNG after placing a call.
I just run a test with my ZyXEL 2864 and it emits CNG after
placing a call regardless of setting AT+FAA=0 or AT+FAA=1.

	matthias
    
    Do this and it will work in many home pbxs. There are still some, 
    that for reasons unknown to me confuse the modem into getting a busy 
    tone and it just hangs up when they answer. Does anybody have any 
    solutions for this?
    
    Good luck
    
    Julio

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Comments: Authenticated sender is <spine001@gold.tc.umn.edu>
From: Julio C Spinelli <spine001@gold.tc.umn.edu>
To: Matthias.Apitz@SOFTCON.de
Date: Wed, 5 Nov 1997 19:12:24 +0000
Subject: Re: flexfax: problem sending faxes to pxb
CC: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

Uuuups, 

Matthias is right, I confused the parameters. and I suppose that he 
is also right this may be modem dependent. I just tried this with USR 
and this was the problem. While in adaptive answer it tries to listen 
to the other side tone to determine whether it is a fax machine or a 
modem and it does not emit the tone.

Julio
> From:          Matthias.Apitz@SOFTCON.de
> Subject:       Re: flexfax: problem sending faxes to pxb
> To:            Julio C Spinelli <spine001@gold.tc.umn.edu> (Julio C Spinelli)
> Date:          Sat, 1 Nov 1997 18:21:35 +0100 (MEZ)
> Cc:            flexfax@sgi.com
> Reply-to:      Matthias.Apitz@SOFTCON.de (Matthias Apitz)

> Julio C Spinelli wrote:
>     
>     To have the modem emit this tone, you need to have autoanswer off, 
>     something like AT+FAA=0 works in many modems. To have this be part of 
>     the hylafax initialization, you need to add a statement similar 
>     (please check the manual pages for setup for the exact spelling) to
>     
>     AdaptiveAnswer: AT+FAA=0 #turns adaptive answer off
>     
>     To your config.xxx file (xxx represents the device name where the 
>     modem is) in the /var/spool/fax/etc directory.
> 
> AdaptiveAnswer: controls as a boolean value whether or not an
> adaptive answer strategy is used (and defaults to No). It has
> nothing to do with the CNG tone and setting AT+FAA=0 into
> this configuration parameter would no work. The problem of
> the original question was also how to make the modem emit CNG
> during a call. AT+FAA=1 is typically set in ModemSetupAACmd:
> to make HylaFAX service a fax, data, or voice call
> as identified by the modem(!). Naturally this will only work if the
> supports a good adaptive-answering facility (for more infos
> see html/setup-advanced.html and the man page config(4F)).
> 
> I don't know if or if not AT+FAA=0 in ModemSetupAACmd: for the
> modem (NB: the original poster didn't give much information
> about the modem) will switch on CNG after placing a call.
> I just run a test with my ZyXEL 2864 and it emits CNG after
> placing a call regardless of setting AT+FAA=0 or AT+FAA=1.
> 
> 	matthias
>     
>     Do this and it will work in many home pbxs. There are still some, 
>     that for reasons unknown to me confuse the modem into getting a busy 
>     tone and it just hangs up when they answer. Does anybody have any 
>     solutions for this?
>     
>     Good luck
>     
>     Julio
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: thoth@cairo.purplefrog.com
To: flexfax@sgi.com
Subject: flexfax: HylaFAX -vs- Boca
Date: Sun, 02 Nov 1997 18:17:05 -0500
Sender: owner-flexfax@celestial.com


  I'm trying to get HylaFAX to send pages (not fax pages, alphanumeric
beeper pages).  However, the faxq process is smoking crack and giving me


Nov  2 18:13:03 cairo PageSend[15600]: PAGE FAILED: Unknown problem (check modem power)

However, Everything seems to go smoothly up to the point where HylaFAX
loses its mind.

Nov 02 18:08:11.76: [15585]: SESSION BEGIN 00000007 13524120001
Nov 02 18:08:11.78: [15585]: SEND PAGE: JOB 3 DEST 412-0001 COMMID 00000007
Nov 02 18:08:11.78: [15585]: DELAY 2600 ms
Nov 02 18:08:14.40: [15585]: <-- [17:ATE0V1Q0X3S0=0H0\r]
Nov 02 18:08:14.54: [15585]: --> [16:ATE0V1Q0X3S0=0H0]
Nov 02 18:08:14.54: [15585]: --> [2:OK]
Nov 02 18:08:14.54: [15585]: <-- [20:ATS8=2S7=60&K&D3&C1\r]
Nov 02 18:08:14.69: [15585]: --> [2:OK]
Nov 02 18:08:14.69: [15585]: <-- [7:ATL1M1\r]
Nov 02 18:08:14.83: [15585]: --> [2:OK]
Nov 02 18:08:14.84: [15585]: <-- [12:AT+FCLASS=0\r]
Nov 02 18:08:15.08: [15585]: --> [2:OK]
Nov 02 18:08:15.08: [15585]: DIAL 4120001
Nov 02 18:08:15.08: [15585]: <-- [12:ATDT4120001\r]
Nov 02 18:08:30.89: [15585]: --> [12:CARRIER 1200]
Nov 02 18:08:30.91: [15585]: PAGE FAILED: Unknown problem (check modem power)
Nov 02 18:08:30.92: [15585]: <-- [5:ATH0\r]
Nov 02 18:08:32.05: [15585]: --> [10:NO CARRIER]
Nov 02 18:08:32.05: [15585]: MODEM No carrier
Nov 02 18:08:32.06: [15585]: SESSION END


  I tried typing all that junk into Seyon and after dialing I get
this:

CARRIER 1200

PROTOCOL: LAP-M

COMPRESSION: V.42BIS

CONNECT 9600/ARQ


   Why can't HylaFAX wait for the CONNECT string?  Mail clues to
thoth@purplefrog.com.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: HylaFAX -vs- Boca
To: thoth@cairo.purplefrog.com
Date: Mon, 3 Nov 1997 15:00:48 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

thoth@cairo.purplefrog.com wrote:
    
    
      I'm trying to get HylaFAX to send pages (not fax pages, alphanumeric
    beeper pages).  However, the faxq process is smoking crack and giving me
    
    
    Nov  2 18:13:03 cairo PageSend[15600]: PAGE FAILED: Unknown problem (check modem power)
    
    However, Everything seems to go smoothly up to the point where HylaFAX
    loses its mind.
    
    Nov 02 18:08:11.76: [15585]: SESSION BEGIN 00000007 13524120001
    Nov 02 18:08:11.78: [15585]: SEND PAGE: JOB 3 DEST 412-0001 COMMID 00000007
    Nov 02 18:08:11.78: [15585]: DELAY 2600 ms
    Nov 02 18:08:14.40: [15585]: <-- [17:ATE0V1Q0X3S0=0H0\r]
    Nov 02 18:08:14.54: [15585]: --> [16:ATE0V1Q0X3S0=0H0]
    Nov 02 18:08:14.54: [15585]: --> [2:OK]
    Nov 02 18:08:14.54: [15585]: <-- [20:ATS8=2S7=60&K&D3&C1\r]
    Nov 02 18:08:14.69: [15585]: --> [2:OK]
    Nov 02 18:08:14.69: [15585]: <-- [7:ATL1M1\r]
    Nov 02 18:08:14.83: [15585]: --> [2:OK]
    Nov 02 18:08:14.84: [15585]: <-- [12:AT+FCLASS=0\r]
    Nov 02 18:08:15.08: [15585]: --> [2:OK]
    Nov 02 18:08:15.08: [15585]: DIAL 4120001
    Nov 02 18:08:15.08: [15585]: <-- [12:ATDT4120001\r]
    Nov 02 18:08:30.89: [15585]: --> [12:CARRIER 1200]
    Nov 02 18:08:30.91: [15585]: PAGE FAILED: Unknown problem (check modem power)
    Nov 02 18:08:30.92: [15585]: <-- [5:ATH0\r]
    Nov 02 18:08:32.05: [15585]: --> [10:NO CARRIER]
    Nov 02 18:08:32.05: [15585]: MODEM No carrier
    Nov 02 18:08:32.06: [15585]: SESSION END
    
    
      I tried typing all that junk into Seyon and after dialing I get
    this:
    
    CARRIER 1200
    
    PROTOCOL: LAP-M
    
    COMPRESSION: V.42BIS
    
    CONNECT 9600/ARQ
    
    
       Why can't HylaFAX wait for the CONNECT string?  Mail clues to
    thoth@purplefrog.com.

Check ClassModem::isNoise() in faxd/ClassModem.c++ to see how
to ignore this noise of the modem;

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 31 Oct 1997 18:56:34 +0100 (CET)
From: "Lorenzo M. Catucci" <lorenzo@argon.roma2.infn.it>
To: flexfax@sgi.com
Subject: flexfax: faxgetty as a callback server.
Sender: owner-flexfax@celestial.com

Dear readers,
	here I am, after an hylafax reinstall, with a question I tryied to
pose a long time ago. Hope I'll not get too many bad answers...

In my experience, faxgetty handles the modem much better than any other
getty I tried, especially in the useful mixed fax/data environment.

Now, mgetty have two nice features, which are autoppp, and callback.
autoppp means I don't have to set up a ppp account, with a common
password, but rather have mgetty recognize ppp traffic, start pppd with
the login and pap options, and all is done with the standard username and
password. Callback means... you know that. Now, I have managed to have
mgetty called from faxgetty, but doing that I lost the callback option.
What I think is needed is a small external program which asks faxgetty (or
faxq, I still haven't understood which program does handle the calling)
via the FIFO and asks it to call a telephone number, and then pass the
serial line to another program, which, should be (I know, you are
surprised) mgetty. Is it feasible? How many people could like such a
feature?

How many could help a mostly fortran/awk coder getting it working?

Yours,


lorenzo

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Sat, 1 Nov 1997 17:31:00 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: "Lorenzo M. Catucci" <lorenzo@argon.roma2.infn.it>
Cc: flexfax@sgi.com
Subject: Re: flexfax: faxgetty as a callback server.
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Fri, 31 Oct 1997, Lorenzo M. Catucci wrote:

> Now, mgetty have two nice features, which are autoppp, and callback.
> autoppp means I don't have to set up a ppp account, with a common
> password, but rather have mgetty recognize ppp traffic, start pppd with
> the login and pap options, and all is done with the standard username and
> password. Callback means... you know that. Now, I have managed to have

Hmmm. Under SunOS, I set up distinct PPP accounts for my users. This
permits me to log, via utmp, when they call in as themselves and when
they call in as PPP users.

There are a gazillion ways to do this, of course....

> mgetty called from faxgetty, but doing that I lost the callback option.
> What I think is needed is a small external program which asks faxgetty (or
> faxq, I still haven't understood which program does handle the calling)
> via the FIFO and asks it to call a telephone number, and then pass the
> serial line to another program, which, should be (I know, you are
> surprised) mgetty. Is it feasible? How many people could like such a
> feature?

What you want, I think, is to have a shell script as your dialup login
shell: this shell would then run off and start up the callback feature.

How do you want to set up what phone number to call back? Typed by user,
or what?

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNFutlz/+ItycgIJRAQFN9AP8DJtPzYs6TwSN5iyX2xolTCXhY6OICWaR
ICC7qARmcfpSbEsLRcqSWsJ/Q4IQ8LWFrDDM7DgGt+6QSnLQQXC7ZtVzaHH7Fh/T
FdoCPxQaoKX1BTODTjue3IHg1LjmcaG6blbHGLE5o6XeUvTt1hbTSk4FdWBuQg6e
K43JRpxcMXc=
=sl9q
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: faxgetty as a callback server.
To: raoul@cirl.meei.harvard.edu (Nico Garcia)
Date: Sun, 2 Nov 1997 08:18:02 +0100 (MEZ)
Cc: lorenzo@argon.roma2.infn.it (Lorenzo M. Catucci), flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Nico Garcia wrote:
    
    On Fri, 31 Oct 1997, Lorenzo M. Catucci wrote:
    
    > Now, mgetty have two nice features, which are autoppp, and callback.
    > autoppp means I don't have to set up a ppp account, with a common
    > password, but rather have mgetty recognize ppp traffic, start pppd with
    > the login and pap options, and all is done with the standard username and
    > password. Callback means... you know that. Now, I have managed to have
    
    Hmmm. Under SunOS, I set up distinct PPP accounts for my users. This
    permits me to log, via utmp, when they call in as themselves and when
    they call in as PPP users.
    
    There are a gazillion ways to do this, of course....

Yes, but the question was not to *login* into the system.
What Lorenzo wants (and what I'ld call a good new feature
for HylaFAX's faxgetty(1M)) is that faxgetty(1M) accepts
a call, checks if it is a data call and checks also if
PPP-packets are coming in from the caller and (if so) directly
launches a PPP-daemon rather than the normal system getty or
ttymon. The problem with that is that faxgetty(1M) has to
read a few chars from the modem to check if they are PPP-data
and if not has to hand-over these characters to the normal
system's getty (e.g. push them back to modem fd to let
getty read them which isn't possible without integrating
the getty into faxgetty(1M)). With PPP the authentication
could be done directly using PAP or CHAP.

On my system at home I've one faxgetty(1M) running which launches
directly the PPP-daemon (instead of ttymon). I simply
patched the name "/usr/lib/saf/ttymon" to "/usr/local/ppp/pppd"
in a copy of sbin/faxgetty. The "modem" on this line is a
ZyXEL 2864I ISDN-TA and the Cisco 2503 in my company uses
CHAP to dial-in via PPP into my system at home. This line
can't be used for "normal" logins of course.

	matthias
    
    > mgetty called from faxgetty, but doing that I lost the callback option.
    > What I think is needed is a small external program which asks faxgetty (or
    > faxq, I still haven't understood which program does handle the calling)
    > via the FIFO and asks it to call a telephone number, and then pass the
    > serial line to another program, which, should be (I know, you are
    > surprised) mgetty. Is it feasible? How many people could like such a
    > feature?
    
    What you want, I think, is to have a shell script as your dialup login
    shell: this shell would then run off and start up the callback feature.
    
    How do you want to set up what phone number to call back? Typed by user,
    or what?
    
    			Nico Garcia
    			Engineer, CIRL 
    			Mass. Eye and Ear Infirmary
    			raoul@cirl.meei.harvard.edu

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Leslie Mikesell <les@Mcs.Net>
Subject: Re: flexfax: faxgetty as a callback server.
To: Matthias.Apitz@SOFTCON.de
Date: Sun, 2 Nov 1997 14:17:09 -0600 (CST)
Cc: raoul@cirl.meei.harvard.edu, lorenzo@argon.roma2.infn.it, flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> 
> Nico Garcia wrote:
>     
>     On Fri, 31 Oct 1997, Lorenzo M. Catucci wrote:
>     
>     > Now, mgetty have two nice features, which are autoppp, and callback.
>     > autoppp means I don't have to set up a ppp account, with a common
>     > password, but rather have mgetty recognize ppp traffic, start pppd with
>     > the login and pap options, and all is done with the standard username and
>     > password. Callback means... you know that. Now, I have managed to have
>     
>     Hmmm. Under SunOS, I set up distinct PPP accounts for my users. This
>     permits me to log, via utmp, when they call in as themselves and when
>     they call in as PPP users.
>     
>     There are a gazillion ways to do this, of course....
> 
> Yes, but the question was not to *login* into the system.
> What Lorenzo wants (and what I'ld call a good new feature
> for HylaFAX's faxgetty(1M)) is that faxgetty(1M) accepts
> a call, checks if it is a data call and checks also if
> PPP-packets are coming in from the caller and (if so) directly
> launches a PPP-daemon rather than the normal system getty or
> ttymon. The problem with that is that faxgetty(1M) has to
> read a few chars from the modem to check if they are PPP-data
> and if not has to hand-over these characters to the normal
> system's getty (e.g. push them back to modem fd to let
> getty read them which isn't possible without integrating
> the getty into faxgetty(1M)). With PPP the authentication
> could be done directly using PAP or CHAP.

There have been some postings to the list about success with letting
faxgetty start mgetty with auto_ppp enabled but it sounds tricky.
If you are using speed-buffering modems and the call is already
answered getty's role is reduced to collecting the first line
of input, changing the line state in utmp, and exec'ing login
with the input line given as an argument to login.  This sounds
to me like a couple of calls to functions that are already in
faxgetty (but things always sound easy if you aren't doing it
yourself...).

Is someone interested in building this functionality into faxgetty
along with the autoppp detection?  We'd save the trouble exec'ing
a 30K program to let mgetty do this tiny bit of work, and more
importantly we could simplify installation because you would no
longer need to build the gettydefs entry to match the faxgetty
modem setup or figure out you local getty's arguments.

  Les Mikesell
    les@mcs.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 3 Nov 1997 11:04:24 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: "Lorenzo M. Catucci" <lorenzo@argon.roma2.infn.it>
Cc: Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: faxgetty as a callback server.
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Sun, 2 Nov 1997, Lorenzo M. Catucci wrote:

> 
> > 
> > Hmmm. Under SunOS, I set up distinct PPP accounts for my users. This
> > permits me to log, via utmp, when they call in as themselves and when
> > they call in as PPP users.
> > 
> I still prefer my users not to try remembering two passwords instead of
> one..., and I really like the pap and login options of ppp

OK. There are half a dozen ways to do this. One simple one is to wrap
a shell script into the standard login, that detects if the serial
line being used for the login is a dialup line. This would then start
the ppp-login script as needed, and wouldn't start it if you were on a
standard dialup line.

The simplest, least obtrusive way to detect *that* is to set the
terminal entries in /etc/ttytab for something like "dialup" instead of
the more common vt100.

> > How do you want to set up what phone number to call back? Typed by user,
> > or what?
> > 
> The simple callback from mgetty takes some options like the serial line to
> open (but this would be unnecessary with faxgetty, and instead could be
> superceded by a 'best available' or some othe policy) and the telephone
> number to call. The nice thing is that mgetty does have a config table
> where could be added lines like (I forgot autoppp was near to my own
> callback, and put both in)

> The best option to have these two programs integrated would then be to be
> able to define a 'hylacallback' program, with similar options to mgetty's,
> and then have faxgetty handle the modem, and mgetty the data connections.
> (I know, this could seem unfair to mgetty's author, but I'm a fan of
> Hylafax fax handling...)

Considering the more flexible and powerful faxgetty systems, I think this
is completely reasonable

The last couple of notes would be of general use to the list, I think.
Go ahead and post, and I'll post this one.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNF31/T/+ItycgIJRAQG/xQP/Y3imkqOUFa4zFsKVXw2DTKyEeB2MXAE4
r22CBZ2kwza8XokdDJ0hMhtrqKWk47LAh0qJTdL0lPAXdIHUGlXJFmsriSc6Cgtu
SVlfNW3FwL+Samk6h5KNzyNLUQmVCinwzzG4PXldbOvOZ0cZ4wj3li/BpIlaZs+T
o9joV8jL8iU=
=j8CL
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 3 Nov 1997 11:15:36 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Matthias Apitz <Matthias.Apitz@SOFTCON.de>
Cc: "Lorenzo M. Catucci" <lorenzo@argon.roma2.infn.it>, flexfax@sgi.com
Subject: Re: flexfax: faxgetty as a callback server.
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Sun, 2 Nov 1997 Matthias.Apitz@SOFTCON.de wrote:

> Nico Garcia wrote:
>     
>     On Fri, 31 Oct 1997, Lorenzo M. Catucci wrote:
>     
>     > Now, mgetty have two nice features, which are autoppp, and callback.
>     > autoppp means I don't have to set up a ppp account, with a common
>     > password, but rather have mgetty recognize ppp traffic, start pppd with
>     > the login and pap options, and all is done with the standard username and
>     > password. Callback means... you know that. Now, I have managed to have
>     
>     Hmmm. Under SunOS, I set up distinct PPP accounts for my users. This
>     permits me to log, via utmp, when they call in as themselves and when
>     they call in as PPP users.
>     
>     There are a gazillion ways to do this, of course....
> 
> Yes, but the question was not to *login* into the system.
> What Lorenzo wants (and what I'ld call a good new feature
> for HylaFAX's faxgetty(1M)) is that faxgetty(1M) accepts
> a call, checks if it is a data call and checks also if
> PPP-packets are coming in from the caller and (if so) directly
> launches a PPP-daemon rather than the normal system getty or
> ttymon. The problem with that is that faxgetty(1M) has to
> read a few chars from the modem to check if they are PPP-data
> and if not has to hand-over these characters to the normal
> system's getty (e.g. push them back to modem fd to let
> getty read them which isn't possible without integrating
> the getty into faxgetty(1M)). With PPP the authentication
> could be done directly using PAP or CHAP.

Fair enough, but this completely replaces the getty normally
associated with a dial-up line with a pppd. That means that that line
can no longer accept simple serial line communications such as Kermit,
Hyperterminal, White Knight, UUCP, etc. That's why I think it's much
more flexible to set this up as a configuration per-user, instead of as
a per modem-line setup. 

It might be set up as some options traditionally are, where if you
don't interrupt it in 5 seconds it falls through and gives you a PPP
access or, in Lorenzo's case, a dialback....

> On my system at home I've one faxgetty(1M) running which launches
> directly the PPP-daemon (instead of ttymon). I simply
> patched the name "/usr/lib/saf/ttymon" to "/usr/local/ppp/pppd"
> in a copy of sbin/faxgetty. The "modem" on this line is a
> ZyXEL 2864I ISDN-TA and the Cisco 2503 in my company uses
> CHAP to dial-in via PPP into my system at home. This line
> can't be used for "normal" logins of course.

Exactly. I think a standard SunOS serial line is more useful having
straight shell capability, to permit people to read their email with
something simple like Kermit instead of futzing with their calling
machine's network setups and using PPP.

>     What you want, I think, is to have a shell script as your dialup login
>     shell: this shell would then run off and start up the callback feature.

Except that Lorenzo wants to have the same account names for his users
as for their dialup use: this makes replacing their login shells
problematic at best. Better to have some sort of option called from
their .login or .profile file, something to then fire off the callback
feature.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNF34oD/+ItycgIJRAQGYLQP/XH7kA0iMIt8Y6bZfNgJdyVNLJ7atEo+I
v5VKGAlCO9hH62WRPR/SXFyfx7eRC8iK2MagS6ODH2L8MTIK5DQb7bRSYsKllJ+7
q8svqBXK+E1BH1NpwE7iFH/xwCxmYHpPG69ELma18L3GmIWEUTibt2x4OpifKnI2
JGcKbGf7uPw=
=03N1
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Tue, 4 Nov 1997 18:42:49 +0100 (MET)
From: Thomas Tornblom <Thomas.Tornblom@Hax.SE>
Reply-To: Thomas Tornblom <Thomas.Tornblom@Hax.SE>
Subject: Re: flexfax: faxgetty as a callback server.
To: Matthias.Apitz@SOFTCON.de, raoul@cirl.meei.harvard.edu
Cc: lorenzo@argon.roma2.infn.it, flexfax@sgi.com
Sender: owner-flexfax@celestial.com


> > 
> > Yes, but the question was not to *login* into the system.
> > What Lorenzo wants (and what I'ld call a good new feature
> > for HylaFAX's faxgetty(1M)) is that faxgetty(1M) accepts
> > a call, checks if it is a data call and checks also if
> > PPP-packets are coming in from the caller and (if so) directly
> > launches a PPP-daemon rather than the normal system getty or
> > ttymon. The problem with that is that faxgetty(1M) has to
> > read a few chars from the modem to check if they are PPP-data
> > and if not has to hand-over these characters to the normal
> > system's getty (e.g. push them back to modem fd to let
> > getty read them which isn't possible without integrating
> > the getty into faxgetty(1M)). With PPP the authentication
> > could be done directly using PAP or CHAP.
> 
> Fair enough, but this completely replaces the getty normally
> associated with a dial-up line with a pppd. That means that that line
> can no longer accept simple serial line communications such as Kermit,
> Hyperterminal, White Knight, UUCP, etc. That's why I think it's much
> more flexible to set this up as a configuration per-user, instead of as
> a per modem-line setup. 
> 

I believe you miss the point. The meaning was to check if it is a PPP 
connection and in that case start a ppp daemon. Otherwise the normal 
getty/ttymon would be called.

I think this would be a nice addition. I have been thinking in these terms 
myself as I have an ISDN TA (ZyXEL 2864I) that accepts both synchronous and 
asynchronous connections and I can't call in from a Sun with an ISDN board as 
there is no provision for chatscripts when running synch PPP. Using this scheme 
they would authenticate using PAP or CHAP.

> It might be set up as some options traditionally are, where if you
> don't interrupt it in 5 seconds it falls through and gives you a PPP
> access or, in Lorenzo's case, a dialback....
> 
> > On my system at home I've one faxgetty(1M) running which launches
> > directly the PPP-daemon (instead of ttymon). I simply
> > patched the name "/usr/lib/saf/ttymon" to "/usr/local/ppp/pppd"
> > in a copy of sbin/faxgetty. The "modem" on this line is a
> > ZyXEL 2864I ISDN-TA and the Cisco 2503 in my company uses
> > CHAP to dial-in via PPP into my system at home. This line
> > can't be used for "normal" logins of course.
> 
> Exactly. I think a standard SunOS serial line is more useful having
> straight shell capability, to permit people to read their email with
> something simple like Kermit instead of futzing with their calling
> machine's network setups and using PPP.
> 
> >     What you want, I think, is to have a shell script as your dialup login
> >     shell: this shell would then run off and start up the callback feature.
> 
> Except that Lorenzo wants to have the same account names for his users
> as for their dialup use: this makes replacing their login shells
> problematic at best. Better to have some sort of option called from
> their .login or .profile file, something to then fire off the callback
> feature.
> 
> 			Nico Garcia
> 			Engineer, CIRL 
> 			Mass. Eye and Ear Infirmary
> 			raoul@cirl.meei.harvard.edu
> 
> -----BEGIN PGP SIGNATURE-----
> Version: 2.6.2
> 
> iQCVAwUBNF34oD/+ItycgIJRAQGYLQP/XH7kA0iMIt8Y6bZfNgJdyVNLJ7atEo+I
> v5VKGAlCO9hH62WRPR/SXFyfx7eRC8iK2MagS6ODH2L8MTIK5DQb7bRSYsKllJ+7
> q8svqBXK+E1BH1NpwE7iFH/xwCxmYHpPG69ELma18L3GmIWEUTibt2x4OpifKnI2
> JGcKbGf7uPw=
> =03N1
> -----END PGP SIGNATURE-----
> 
> 

Real life:      Thomas Trnblom             Email:  Thomas.Tornblom@Hax.SE
Snail mail:     HB Hax                      Phone:    +46  18 290 290
                Banvallsvgen 14            Fax:      +46  18 290 291
                S - 754 40 Uppsala, Sweden  Cellular: +46 708 323118

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Sun, 2 Nov 1997 22:07:04 +0100 (CET)
From: "Lorenzo M. Catucci" <lorenzo@argon.roma2.infn.it>
Reply-To: "Lorenzo M. Catucci" <lorenzo@argon.roma2.infn.it>
To: flexfax@sgi.com
Subject: flexfax: My proposal for flexfax
Sender: owner-flexfax@celestial.com

Now,
	since I see there is some interest, I follow up to the thread...
At first, I am sure the whole dirty work could be well passed to faxgetty,
which could then supercede mgetty, but I see mgetty works well, if it is
configured to handle a direct data line which was handed to it from
faxgetty, and I think we could well continue with such a configuration.
I'm sorry, I don't remember the details, which were on a system stolen by
someone in the beginning of september, with all my mails, but I remember
the most imoprtan clue came from a people working at SuSe un Germany.
My real question was about a callback capability, either autonomous, or
configured like in mgetty's login.config lines I have sent yesterday to
Nico:

cback-lore      -       -       /usr/sbin/callback -S 123456789
#
#   New comment:                                  ^^^^ this option
#						tells mgetty to use
#						the same line the call
#						came from; the last
#						one is the telephone
#						number to call
#
# Automatic PPP startup on receipt of LCP configure request (AutoPPP).
#  Consult the "pppd" man page to find pppd options that work for you.
#
#  NOTE: for *some* users, the "-detach" option has been necessary, for 
#        others, not at all. If your pppd doesn't die after hangup, try
it.
#
/AutoPPP/ -     -       /usr/sbin/pppd auth -chap +pap login modem -detach

--------------------------------------------------------------------------
Now, what is missing, IMHO, is a callback capability for faxgatty/hfaxd,
since the data handling by mgetty is really nice, with options for spawing
autoUUCP, autoFIDO and autoPPP, or an external callback program. Now, 
mgetty's one does wirk by calling directly mgetty, which should then 'own'
the modem line, and be able to talk to the modem, and which I'd rather
like to be superceded by faxgetty.

Yours,


lorenzo

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
X-Authentication-Warning: guinness.chelsea.net: davidb owned process doing -bs
Date: Mon, 3 Nov 1997 13:22:22 -0500 (EST)
From: David Birnbaum <davidb@chelsea.net>
Reply-To: David Birnbaum <davidb@chelsea.net>
To: flexfax@sgi.com
Subject: flexfax: DialRules - external variable input
Sender: owner-flexfax@celestial.com

Folks,

I was perusing the DialRules.c++ code to determine if it was possible to
insert a variable that is determined dynamically from the environment or
the fax ID #.  Basically, I want to have a rule something like this:

  ^	= #${serial},

where the $serial variable is a serial number I send prior to dialing the
number.  The $serial could be the commid or the jobid number.

The reason I would like to do this is that we get a feed directly from our
phone switch with pricing information and such, but there's no sure-fire
way to link a phone switch record to a fax transfer record.  (yes, you can
get close, but there's the occasional record which fails to match up,
espcially when we do big broadcast faxes, and it's a nightmare if the
times drift by more than a few seconds).

Ideally, by sending a PIN code that's actually a serial number, I am in
great shape to do matching.  So, for job 1523, the number "5551212" would
become "1523#,5551212", my phone record would contain PIN 1523, and I
would be able to match it to job 1523 and do my entitlement billing.

Any thoughts?

Thanks,

David.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Tue, 04 Nov 1997 20:33:29 +0100
Reply-To: spilker@t-online.de
Organization: JetSys
To: flexfax@sgi.com
Subject: flexfax: Phase D/C error when sending a fax
From: spilker@t-online.de (Jrg Spilker)
Sender: owner-flexfax@celestial.com

Hello,

ive some problems with sending a fax with the hylafax package
(receiving works fine with all kinds of modems and fax machines on the
other side).

My modem is a Elink 343 ISDN TA. It should comply to Fax Class 2
according to the docs. Its connected to a EuraCom 181 PBX which is
connected to an ISDN BRI.

Hylafax is version 4.0pl1 on a Linux machine with Kernel 2.0.30

The log shows always something like this

Oct 26 08:03:05.23: [25906]: DIAL 005222986899
Oct 26 08:03:05.23: [25906]: DELAY 200 ms
Oct 26 08:03:18.02: [25906]: REMOTE CSI "+49 5222 986899"
Oct 26 08:03:18.32: [25906]: REMOTE best rate 9600 bit/s
Oct 26 08:03:18.32: [25906]: REMOTE max page width 1728 pixels in 215 mm
Oct 26 08:03:18.32: [25906]: REMOTE max unlimited page length
Oct 26 08:03:18.32: [25906]: REMOTE best vres 7.7 line/mm
Oct 26 08:03:18.32: [25906]: REMOTE best format 2-D MR
Oct 26 08:03:18.32: [25906]: REMOTE best 20 ms, 10 ms/scanline
Oct 26 08:03:18.32: [25906]: USE 9600 bit/s
Oct 26 08:03:18.32: [25906]: USE 40 ms/scanline
Oct 26 08:03:18.32: [25906]: SEND file "docq/doc14.ps;01"
Oct 26 08:03:18.42: [25906]: USE page width 1728 pixels in 215 mm
Oct 26 08:03:18.42: [25906]: USE unlimited page length
Oct 26 08:03:18.42: [25906]: USE 7.7 line/mm
Oct 26 08:03:18.42: [25906]: USE 1-D MR
Oct 26 08:03:18.42: [25906]: DELAY 200 ms
Oct 26 08:03:18.94: [25906]: DELAY 200 ms
Oct 26 08:03:25.64: [25906]: SEND wait for XON
Oct 26 08:03:25.68: [25906]: SEND begin page
Oct 26 08:04:03.69: [25906]: SENT 49738 bytes of data
Oct 26 08:04:03.69: [25906]: SEND end page
Oct 26 08:04:07.54: [25906]: SEND send MPS (more pages, same document)
Oct 26 08:04:07.54: [25906]: DELAY 200 ms
Oct 26 08:04:10.33: [25906]: MODEM protocol botch ("+FPTS:"), can not
parse PPR

and i receive allways this mail.

Your facsimile job to 005222986899 was not sent because:
    Communication failure during Phase B/C

The receiver was a fax machine on the other side. This was one of the
better results. In most cases the connect to the other side was
immediately terminated after the initial handshake with the "Com failre
during B/C" error.

Do you have any idea why there are these problems? Before changing to
Hylafax, i used the sendfax part from the mgetty+sendfax package without
any problems.

Greetings, Joerg

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Tue, 04 Nov 1997 13:33:42 -0600
To: flexfax@sgi.com
From: Greg Nicholson <greg@d0gz.net>
Subject: flexfax: WHY?  :)  facsimile not received
Sender: owner-flexfax@celestial.com

	Greetings all,

		I've been having problems receiving faxes recently with the following
transcript as the reason why.  I don't understand why it's not an
acceptable TSI, and what do I need to do in order to change HylaFAX to
recieve this type?

	Modem is a USR 28.8 VI... I know, not the best one to be using, in FCLASS
2.0.

HylaFAX Version 4.0pl1 running on Redhat 4.2

Thanks!

Greg Nicholson

   --------------------------------------
An attempt to receive facsimile on ttyS1 failed because:

 

 ---- Transcript of session follows ----

Nov 04 13:24:23.26: [ 286]: SESSION BEGIN 00000204 16157943696
Nov 04 13:24:23.26: [ 286]: <-- [4:ATA\r]
Nov 04 13:24:30.55: [ 286]: --> [4:+FCO]
Nov 04 13:24:30.55: [ 286]: ANSWER: FAX CONNECTION
Nov 04 13:24:30.56: [ 286]: RECV FAX: begin
Nov 04 13:24:32.21: [ 286]: --> [27:+FTI:" 6174407072"]
Nov 04 13:24:32.21: [ 286]: REMOTE TSI "6174407072"
Nov 04 13:24:32.49: [ 286]: --> [20:+FCS:0,5,0,2,0,0,0,0]
Nov 04 13:24:32.49: [ 286]: REMOTE wants 14400 bit/s
Nov 04 13:24:32.49: [ 286]: REMOTE wants page width 1728 pixels in 215 mm
Nov 04 13:24:32.50: [ 286]: REMOTE wants unlimited page length 
Nov 04 13:24:32.50: [ 286]: REMOTE wants 3.85 line/mm
Nov 04 13:24:32.50: [ 286]: REMOTE wants 1-D MR
Nov 04 13:24:32.61: [ 286]: --> [2:OK]
Nov 04 13:24:32.62: [ 286]: REJECT TSI "6174407072"
Nov 04 13:24:32.63: [ 286]: RECV FAX (00000204): recvq/fax00004.tif from
6174407072, route to <unspecified>, 0 pages in 0:02
Nov 04 13:24:32.65: [ 286]: RECV FAX: Permission denied (unacceptable
client TSI)
Nov 04 13:24:32.65: [ 286]: <-- [7:AT+FKS\r]
Nov 04 13:25:02.65: [ 286]: MODEM <Timeout>
Nov 04 13:25:02.66: [ 286]: RECV FAX: bin/faxrcvd "recvq/fax00004.tif"
"ttyS1" "00000204" "" 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
X-Authentication-Warning: tui.pinnacle.co.nz: jonc owned process doing -bs
Date: Wed, 5 Nov 1997 08:42:42 +1300 (NZDT)
From: Jonathan Chen <jonc@pinnacle.co.nz>
To: Ancil Bethelmy <cowpoke@wave.cape.net>
cc: HylaFAX Mailing List <flexfax@sgi.com>
Subject: Re: flexfax: questions about agetty, mgetty and getty
Sender: owner-flexfax@celestial.com

On Mon, 3 Nov 1997, Ancil Bethelmy wrote:

> I have been trying to setup HylaFAX-v4.0pl1 on a DEC alpha.
> 
> Here is the problem. I have been asked by the software during setup
> where the  getty programs agetty and vgetty are. I have been unable to
> get any code to compile for these programs. Any ideas on where to get
> the code for a dec alpha? Failing that, is it possible to use the
> regular getty with the HylaFAX software? Where in the config process do
> you tell it to ignore looking for agetty and vgetty.

agetty and vgetty are optional hooks for non-data/fax calls. You can
supply a non-existent pathnames (the last I looked anyway), and the
system will accept it.
--
Jonathan Chen <jonc@pinnacle.co.nz>                 Once is dumb luck.
                                                 Twice is coincidence.
             Three times and Somebody Is Trying To Tell You Something.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Tue, 4 Nov 1997 17:43:03 -0500 (EST)
From: Chad M Vande Pol <chad@is6.iserv.net>
To: flexfax@sgi.com
Subject: flexfax: Multiple Modems
Sender: owner-flexfax@celestial.com


Is there any thing I need to do to get Hylafax to work over multiple
modems?  I have added modems But Hylafax seems to only send things out one
modem at a time.  I can send 3 faxes and 4 alpha pages but everything just
gets queued up and goes out FIFO though a modem while my other modems just
sit idle.  Is it posible to use more than one modem at a time?  I have
played with modem prioities and stuff but can not seem to figure it out.

Also has anyone been successful using a modem bank such as an Ascend Max
4000 or cisco AS5200, for sending faxes or pages?

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
To: Chad M Vande Pol <chad@is6.iserv.net>
cc: flexfax@sgi.com
Subject: Re: flexfax: Multiple Modems 
Date: Wed, 05 Nov 1997 14:03:29 +0000
From: "Mr. Arlington Hewes" <tpcadmin@info.tpc.int>
Sender: owner-flexfax@celestial.com

>>>>> On Tue, 4 Nov 1997, "CMVP" == Chad M Vande Pol wrote:

  CMVP> Also has anyone been successful using a modem bank such as an Ascend
  CMVP> Max 4000 or cisco AS5200, for sending faxes or pages?

Chad,

Aside from yourself, I did not have any replies to my question about
managing Ascend modems with HylaFAX last week.  I can only assume that
nobody has been able to do this. This is a tremendous shame . . . all those
modems!!!

-Darren

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Wed, 5 Nov 1997 22:21:40 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Karl Feichtinger <lists@comconnect.com>
Cc: flexfax@sgi.com
Subject: Re: flexfax: How best use multiple modems?
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Wed, 5 Nov 1997, Karl Feichtinger wrote:

> Hi,
> 
> I know this topic has been discussed earlier, however, I would be glad to
> receive a specific advice. I want to connect 6 modems to a Linux machine. 
> 
> I wonder whether I should use internal modems (this reduces cabling) or
> external modems, in which case I need a recommendation for a multi-port
> serial board.
> Which minimum RAM and Processor requirements would I need? SCSI?

More than 2 modems can cause grief real fast. PC clones normally have
only two interrupt codes for serial ports. You can't just add new ones
willy-nilly, they're very much built into the hardware and the
drivers.

I suggest that for this many modems, you definitely need some sort of
serial expander. Alternatively, there are several types of network
ready multiple modem servers which might save you a lot of work and
grief and be overall more reliable: you might then be able to use the
Linux box to handle the fax service. I believe some people have done
this on this group....

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNGE3tD/+ItycgIJRAQEMjQP/aqiwNa2G7kDLkaXWOxSt8zX094wPpM7s
C+fvGmmx08jfv8lMtrzDNFA1TqU+kCnID1sSnYv0BKjEX/DX/2ucQndw1gePeRHh
bGj9/fhj9Qd4ElAmprKMj0CYNYeAEi/Q7GxXCBRAzK/y3HWugYj8OCSaMDlPCeZ3
sTRPFkw14rY=
=wwHy
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 06 Nov 1997 08:03:33 -0500 (EST)
Organization: The Creative Group
From: Patrick Gardella <patrick@cre8tivegroup.com>
To: Karl Feichtinger <lists@comconnect.com>
Subject: RE: flexfax: How best use multiple modems?
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

I am running 7 Multitech 2834ZDX MultiModems on a Cyclades 8Y0 multi-port
serial card.  The computer is a P-200 with 128 Megs RAM and 5gig HD.

This computer is overkill, but I wanted it that way :)  Unless we are running
heavily loaded on the fax side (greater than 3,000 faxes waiting in the sendq),
our average load is .15-.2.  I run about 2000 faxes a day in outgoing only.

The OS is FreeBSD 2.2.1.

Since P-200 are reasonably cheap, I'd use one or higher.  32 Megs RAM should be 
fine; 64 Megs better.  Especially if you make your swap big.  Otherwise it 
depends on what all you are going to do with it.  Web server?  Router?

I am very impressed with te Mulitechs, although there was some discussion about
2 weeks ago.  You can check the archives for the deatils.

Patrick Gardella


On 05-Nov-97 Karl Feichtinger wrote:
>Hi,
>
>I know this topic has been discussed earlier, however, I would be glad to
>receive a specific advice. I want to connect 6 modems to a Linux machine. 
>
>I wonder whether I should use internal modems (this reduces cabling) or
>external modems, in which case I need a recommendation for a multi-port
>serial board.
>Which minimum RAM and Processor requirements would I need? SCSI?
>
>Thanks for any advice!
>	Karl 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Comments: Authenticated sender is <spine001@gold.tc.umn.edu>
From: Julio C Spinelli <spine001@gold.tc.umn.edu>
To: flexfax@sgi.com
Date: Wed, 5 Nov 1997 19:12:24 +0000
Subject: flexfax: Problem executing faxrcvd, SOS!
Sender: owner-flexfax@celestial.com

I am running Linux 2.0.29 or something like this. I downloaded the 
hylafax v4pl1 version installed and compiled it as recommended in the 
literature.

My problem is that when I receive a fax, faxrcvd script is not 
executed it bounces with error and nothing happens. I checked the 
code and it is calling /bin/sh sh -c faxrcvd ...... (plus the 
parameters passed to faxrcvd). When I run it from the command line 
(/bin/sh -c faxrcvd .....) it does not work saying that the number of 
parameters is wrong. If I omit the -c it works with no problem in the 
command line.

Does anybody have a clue on what to do?

Why did configure pick up /bin/sh and not /bin/bash?

Can this be the problem? I saw in the text of the configure script 
that Linux can have problems with sh.

Any help will be greatly appreciated,

Julio

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Wed, 5 Nov 1997 14:15:31 -0500 (EST)
From: Evan Leibovitch <evan@telly.org>
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
cc: Mario Misic <mario@laus.hr>, flexfax@sgi.com
Subject: flexfax: Received fax legal/Printer has letter
Sender: owner-flexfax@celestial.com


This is something that doesn't seem *too* difficult, but it would save me
time if someone's already done it:

I've made a one-line change to faxrcvd to automatically print faxes
as they come in, on a postscript printer. Problem is, this printer only
has letter-sized paper and occasionally someone sends a legal-sized fax.

Is there a commonly-used method to detect this and shrink the image to fit
the shorter paper?

Thanks.

------------------------------------------------------------------------------
 Evan Leibovitch, Sound Software Ltd, located in beautiful Brampton, Ontario
Supporting PC-based Unix since 1985 / Caldera & SCO authorized / www.telly.org
----------------- HURD is to Linux as Plan 9 is to System V ------------------

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
X-Authentication-Warning: tui.pinnacle.co.nz: jonc owned process doing -bs
Date: Fri, 7 Nov 1997 08:11:07 +1300 (NZDT)
From: Jonathan Chen <jonc@pinnacle.co.nz>
To: Hylafax Mailing list <flexfax@sgi.com>
Subject: flexfax: Hylafax on BSDI 3.0, running stronghold 2. (fwd)
Sender: owner-flexfax@celestial.com


---------- Forwarded message ----------
Date: Thu, 6 Nov 1997 12:11:35 -0500 (EST)
From: Helen Birkmann <hbirkmann@mkl.com>
To: jonc@pinnacle.co.nz
Subject: Hylafax on BSDI 3.0, running stronghold 2.

I am trying to run hfaxd on this server to use in cgi forms that fax the
input rather than mailing it.  The cgi scripts in question are located on
the same server.

I believe I resolved a previous problem which was a port conflict between 
the SNNP protocol port and the stronghold SSL by redirecting SSL to port
446 instead of 444.

I can now send faxes from the console using sendfax using the faxq with
the faxgetty daemon, but I would like to be able to have faxes processed
when any host on the Internet uses the cgi script.

I still get an error when starting up hfaxd via inetd.conf:  "getpeername
Socket operation on non-socket" and it also tells me that it can't find
local host 4559 if I try to run hfaxd.


So none of the form input gets faxed out.  The documentation is not clear
where to define that any host should be able to use the fax server in this
fashion.  Can you point me in the right direction?

Tx
Helen 
************************************************************************
*	MKL.Net		        	Internet Service Provider      *
*	A service of CPU Industries                                    *
*	Bainbridge, New York, USA      	PPP, dial-up, telnet, uucp,    *
*	http://www.mkl.com		virtual domains, web design    *
*	https://mall.mkl.com	Secured server - Merchant services     *
************************************************************************

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Bug: %z item in ModemFmt broken (hfaxd problem)
To: asparks@nss.harris.com (Alan Sparks)
Date: Fri, 7 Nov 1997 08:30:45 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Alan Sparks wrote:
    
    Found a tiny little bug in hfaxd.  The %z ('*' if faxgetty running) ModemFmt
    format item prints garbage.
    
    Problem is at line 281 of hfaxd/Status.c++, the following fprintf:
    
                 case 'z':
                    fprintf(fd, fspec, config.isGettyRunning ? "*" : " ");
    
    The double-quotes must be changed to single-quotes, since this must match
    the format character selected from the mformat[] table prior to the
    function; at line 207, the format character is set to 'c'.
    
    %z will work OK after this fix.
    -Alan
    
    ---------
    Alan Sparks, IS Engineering Support                   <asparks@harris.com>
    Harris Network Support Systems, Camarillo CA    (805) 389-2430

This will be fixed in pl2; thanks for digging into that anyway.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Leslie Mikesell <les@Mcs.Net>
Subject: Re: flexfax: How best use multiple modems?y
To: raoul@cirl.meei.harvard.edu (Nico Garcia)
Date: Thu, 6 Nov 1997 14:31:04 -0600 (CST)
Cc: lists@comconnect.com, flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> > I know this topic has been discussed earlier, however, I would be glad to
> > receive a specific advice. I want to connect 6 modems to a Linux machine. 
> > 
> > I wonder whether I should use internal modems (this reduces cabling) or
> > external modems, in which case I need a recommendation for a multi-port
> > serial board.
> > Which minimum RAM and Processor requirements would I need? SCSI?
> 
> More than 2 modems can cause grief real fast. PC clones normally have
> only two interrupt codes for serial ports. You can't just add new ones
> willy-nilly, they're very much built into the hardware and the
> drivers.
> 
> I suggest that for this many modems, you definitely need some sort of
> serial expander. Alternatively, there are several types of network
> ready multiple modem servers which might save you a lot of work and
> grief and be overall more reliable: you might then be able to use the
> Linux box to handle the fax service. I believe some people have done
> this on this group....
  
Serial ports are only a problem if you use ordinary PC UART chips
where the CPU does all the work.  An intelligent serial port
expansion card doesn't impose any more load than a network card
moving a similar amount of data.  I've used 6 modems on a 486
running sysvr4 with a digiboard C/X card without problems (I
think ghostscript was more of a bottleneck than the ports), but
you can't get drivers for that card for Linux.  I've used an 
8-port Cyclades card on a Linux box and it seemed fine, but I
never really loaded it.

  Les Mikesell
    les@mcs.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: How best use multiple modems?
To: raoul@cirl.meei.harvard.edu (Nico Garcia)
Date: Fri, 7 Nov 1997 08:20:07 +0100 (MEZ)
Cc: lists@comconnect.com, flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Nico Garcia wrote:
    
    On Wed, 5 Nov 1997, Karl Feichtinger wrote:
    
    > Hi,
    > 
    > I know this topic has been discussed earlier, however, I would be glad to
    > receive a specific advice. I want to connect 6 modems to a Linux machine. 
    > 
    > I wonder whether I should use internal modems (this reduces cabling) or
    > external modems, in which case I need a recommendation for a multi-port
    > serial board.
    > Which minimum RAM and Processor requirements would I need? SCSI?
    
    More than 2 modems can cause grief real fast. PC clones normally have
    only two interrupt codes for serial ports. You can't just add new ones
    willy-nilly, they're very much built into the hardware and the
    drivers.

As long as you have *free* interrupts (you may also use higher
interrupts like 10, 12, 14) you can add serial cards.
I've a 486 PC with 16 MByte RAM at home
which drives 4 serial ports with speeds up to 460.000 bps
without any kind of problem. The PC is running SVR4.2 and runs
a lot of other daemons: HylaFAX's daemons, innd, httpd, nnmaster...
    
    I suggest that for this many modems, you definitely need some sort of
    serial expander. Alternatively, there are several types of network
    ready multiple modem servers which might save you a lot of work and
    grief and be overall more reliable: you might then be able to use the
    Linux box to handle the fax service. I believe some people have done
    this on this group....
    
    			Nico Garcia
    			Engineer, CIRL 
    			Mass. Eye and Ear Infirmary
    			raoul@cirl.meei.harvard.edu

Modems connected to network servers will not work reliable with
HylaFAX because you can't control the time over a TCP/IP link
to match the T.30 timer demands.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Comments: Authenticated sender is <spine001@gold.tc.umn.edu>
From: Julio C Spinelli <spine001@gold.tc.umn.edu>
To: Matthias.Apitz@SOFTCON.de
Date: Fri, 7 Nov 1997 09:03:55 +0000
Subject: Re: flexfax: How best use multiple modems?
CC: lists@comconnect.com, flexfax@sgi.com
Sender: owner-flexfax@celestial.com


> Modems connected to network servers will not work reliable with
> HylaFAX because you can't control the time over a TCP/IP link
> to match the T.30 timer demands.
> 
> 	matthias
> 
> 
Can we use modems that support class 2 Rockwell type command set over 
the TCP/IP network instead of connecting them to the serial devices? 
Would they handle the T.30 timer demands internally?

Julio

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: How best use multiple modems?
To: spine001@gold.tc.umn.edu (Julio C Spinelli)
Date: Fri, 7 Nov 1997 15:39:12 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Julio C Spinelli wrote:
    
    
    > Modems connected to network servers will not work reliable with
    > HylaFAX because you can't control the time over a TCP/IP link
    > to match the T.30 timer demands.
    > 
    > 	matthias
    > 
    > 
    Can we use modems that support class 2 Rockwell type command set over 
    the TCP/IP network instead of connecting them to the serial devices? 
    Would they handle the T.30 timer demands internally?

The modem will handle the timer correctly (if it does handle
them correctly et all) but you can't make sure to pump the
data or AT-cmds to the modem (or read answers from the modem)
in the timer intervalls.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: jay@dixonssurgical.co.uk (Jay Dixon)
To: HylaFAX Mailing List <flexfax@sgi.com>
Subject: flexfax: HylaFAX and PPP on one modem?
Date: Thu, 06 Nov 1997 21:24:42 GMT
Organization: Dixons Surgical Ltd
Reply-To: jay@dixonssurgical.co.uk
Sender: owner-flexfax@celestial.com

Dear All

I am running FreeBSD 2.2.1 with UserPPP to connect my LAN to the net
on demand using the -auto switch and HylaFAX 4.0 pl1 to send faxes
from the LAN.

I am trying to use the same modem (a Hayes Accura 56K External -
Class1 fax) on cuaa0 with both packages.  Is this possible?

I have got both packages working seamlessly when run one at a time but
when I run both together HylaFAX always gives me the 'Waiting for
modem to come ready' message.  Is there a way around this?  Has anyone
got this setup running smoothly?

Thanks
Jay

---------
Jay Dixon - Dixons Surgical Ltd - Wickford - Essex - England
jay@dixonssurgical.co.uk  -  http://www.dixonssurgical.co.uk

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 07 Nov 1997 09:05:08 +0000
To: HylaFAX List <flexfax@sgi.com>
From: Phil Watkinson <pkw@elgro.co.uk>
Subject: Re: flexfax: HylaFAX and PPP on one modem?
Cc: jay@dixonssurgical.co.uk
Sender: owner-flexfax@celestial.com

At 21:24 06/11/1997 GMT, Jay Dixon wrote:
>Dear All
>
>I am running FreeBSD 2.2.1 with UserPPP to connect my LAN to the net
>on demand using the -auto switch and HylaFAX 4.0 pl1 to send faxes
>from the LAN.
>[snip]
>I have got both packages working seamlessly when run one at a time but
>when I run both together HylaFAX always gives me the 'Waiting for
>modem to come ready' message.  Is there a way around this?  Has anyone
>got this setup running smoothly?
>

Hi. I have not come across 'UserPPP' (nor FreeBSD for that matter) but I
suspect your problem relates to lock files.

The html documentation has some useful info of this, and the revised docs
at :

	http://www.elgro.demon.co.uk/hylafax/

have been enhanced in this particular area, mentioning the race condition
that causes the 'Waiting for modem to come ready' message. Please see the

	"faxgetty and getty problems" and
	"problems with uucp, cu, tip, etc"

sections in troubleshooting.html, and the section on "UUCP Lock Files" in
setup-advanced.html.

These updated docs will be included in the v4.0.2 release, but there is a
tarball of all files at the site if you need them now (see index.html).

Good luck,

Phil Watkinson,
Boston, UK.

[ PS : Louis REINARD!  I haven't forgotten about faxmail.c++ and
TextFmt.c++, but I am very busy at work for about another week ].

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 06 Nov 1997 15:25:37 -0600
From: JK Lavacek <jlav@cpicorp.com>
To: flexfax@sgi.com
Subject: flexfax: PCL to postscript?
Sender: owner-flexfax@celestial.com

Has anyone had experience/success with a conversion of pcl formatted
files into postscript?
Seems the format of files recieved onto the faxserver, will be in pcl
format.  Due to things 
beyond my control, I have to live with this.  Attempted to compile a
source distribution of 
a file called "lj2ps".  This gives various errors in the makefile.  With
a limited knowledge 
of C, I'm lost.  Seems the line endind \ is confusing my version of
gcc...? Removed a few \ 
and joined lines in some .c files and the error moved down in lines. 
Any help would be greatly 
appreciated.  Thanks in advance,
			Jason Lavacek <jlav@cpicorp.com>

HYLAfax version:  4.0 patch level 1 
Linux (redhat 4.2) kernel 2.0.30

Jason Lavacek
CPI Corp. 
St. Louis, MO 63103
mailto:jlav@cpicorp.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
X-Authentication-Warning: tui.pinnacle.co.nz: jonc owned process doing -bs
Date: Fri, 7 Nov 1997 13:21:14 +1300 (NZDT)
From: Jonathan Chen <jonc@pinnacle.co.nz>
To: Ancil Bethelmy <cowpoke@wave.cape.net>
cc: HylaFAX Mailing List <flexfax@sgi.com>
Subject: Re: flexfax: What device name for com1 on dec alpha
Sender: owner-flexfax@celestial.com

On Wed, 5 Nov 1997, Ancil Bethelmy wrote:

> This may seem like an unlikely question for this list but if there is
> anyone who has done HylaFAX for a DEC ALPHA please let me know what the
> device setting for a modem on com1 should be. I wonder if it is
> /dev/tty01.

/dev/tty00
--
Jonathan Chen <jonc@pinnacle.co.nz>                 Once is dumb luck.
                                                 Twice is coincidence.
             Three times and Somebody Is Trying To Tell You Something.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 7 Nov 1997 09:47:49 +0000 (GMT)
From: James Dyer <jad@uk.psi.com>
To: flexfax@sgi.com
Subject: flexfax: Multiple Servers
Sender: owner-flexfax@celestial.com

Hi,

I have been tasked with the job of creating a resilient faxing solution
by my employer...

Briefly, the system will have two servers, each at a different physical
location. Each machine will be a SPARC 20, with 8 modems attached to a
Moxa CN2100 Async Server.

Having looked at Hylafax, it appears that persuading it to talk to 8
modems will not be a problem. What I would like to do, however, is create
a pool of the 16 total modems, which will be available for faxing. Should
a network link go down, however, the pool would have to drop back to the 8
local modems. 

My question is simple : Can this be done, and if so, does anyone have any
good pointers on where to start?

J

---
James Dyer  PSINet UK Ltd. SA  work - jad@uk.psi.com play - jad@jad.co.uk
'YOU'RE computer's broken, and I'VE got a problem???'

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 7 Nov 1997 11:09:09 --100
From: coppelli@sun3.dsea.unipi.it (Alessandro Coppelli)
To: flexfax@sgi.com
Subject: flexfax: From tiff to ps and then to MSWord
Sender: owner-flexfax@celestial.com


 Hi to all.
 
 
 I have this problem :
 
 
  I would to want to get one received fax ( format tiff ) and read it then with
  
  MSWord 97 . How I can to do ?.
  
  I have one server fax ( hylafax in Linux 2.0.30 box ) .
  
  I must use tiff2ps ?
  
  and then how I can read it in MSWord 97 ?
  
  
  
  Alessandro

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 07 Nov 1997 11:24:23 -0500
To: Julio C Spinelli <spine001@gold.tc.umn.edu>
From: Aidan Van Dyk <aidan@magma.ca>
Subject: Re: flexfax: Multiple Modems 
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

At 08:48 AM 11/7/97 +0000, you wrote:
>Humm...,
>
>It seems to me that what we would need to do is to write a driver 
>that maps a device to a socket and then run hylafax with the name of 
>that device as the serial device.
>
>Can any Unix guru please indicate if this is the way to go?
>
>If it is, it would be a question of finding a group interested in 
>developing this. I'm willing to help, but I lack the necessary 
>expertise,


I think that this is part of the problem.  Our MAX TNT has about 670 modems
I think (I'm not quite sure, I'ld have to look it up).  I don't thinke we
really want 500+ devices that HylaFAX is talking to as serial ports which
have drivers communicating with the Rack BOX.  We have to find some more
feasable way to do it.

Aidan.
Aidan Van Dyk                        aidan@magma.ca
---------------------------------------------------------------------
Magma Communications Ltd.            Internet Solutions
Unit 210, 52 Antares Drive              for the Ottawa Area
Nepean, ON, K2E 7Z1  CANADA          http://www.magma.ca/

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: /etc/utmp file and faxgetty
To: mig@miso.wwa.com (Miguel Gualdron)
Date: Fri, 7 Nov 1997 16:46:26 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Miguel Gualdron wrote:
    
    I'm having a small problem with utmp and faxgetty.  I have a Linux box with 
    kernel 2.0.30 and Hylafax 4.0pl1.  I have had both inbound and outboud fax 
    services working ok for a couple of months now.  When I get a data call, and 
    the user logs in, both /etc/utmp, and /etc/wtmp are updated correctly.  When 
    the user logs out, the 'last' command reports correctly the length of time the 
    user was logged in, but the 'who' command tells me the user is still logged 
    in.  Apparently faxgetty is not cleaning up the utmp file after a logout.
    
    I have successfully used agetty and getty_ps as the data-side of faxgetty, but 
    the logout problem persists with both of them.
    
    I used to use a sym-link for /dev/ttyS2 called /dev/modem; but I got rid of 
    it, and the problem still exists.
    
    Any help or ideas on this would be appreciated.
    
    --Miguel

Run strace(1) on the faxgetty proc (use also the -f flag to monitor
forked childs) and dial-in. Check the output to find out why
it isn't updating the utmp file. Don't post the strace file
to the list (it contains your typed password).

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
To: flexfax@sgi.com
Subject: flexfax: /etc/utmp file and faxgetty
Date: Fri, 07 Nov 1997 06:52:30 -0600
From: Miguel Gualdron <mig@miso.wwa.com>
Sender: owner-flexfax@celestial.com

I'm having a small problem with utmp and faxgetty.  I have a Linux box with 
kernel 2.0.30 and Hylafax 4.0pl1.  I have had both inbound and outboud fax 
services working ok for a couple of months now.  When I get a data call, and 
the user logs in, both /etc/utmp, and /etc/wtmp are updated correctly.  When 
the user logs out, the 'last' command reports correctly the length of time the 
user was logged in, but the 'who' command tells me the user is still logged 
in.  Apparently faxgetty is not cleaning up the utmp file after a logout.

I have successfully used agetty and getty_ps as the data-side of faxgetty, but 
the logout problem persists with both of them.

I used to use a sym-link for /dev/ttyS2 called /dev/modem; but I got rid of 
it, and the problem still exists.

Any help or ideas on this would be appreciated.

--Miguel

>Received: (from john@localhost) by johncon.johncon.com (8.6.12/8.6.12) id PAA00674; Sat, 8 Nov 1997 15:28:00 -0800
Date: Sat, 8 Nov 1997 15:28:00 -0800
From: John Conover <john@johncon.johncon.com>
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
Cc: John Conover <john@johncon.johncon.com>, flexfax@sgi.com
Subject: Re: flexfax: Cardinal modem.
Reply-To: John Conover <john@johncon.johncon.com>
X-Return-Receipt-To: john@johncon.johncon.com
X-Read-Receipt-To: john@johncon.johncon.com
Sender: owner-flexfax@celestial.com

Nico Garcia writes:
> 
> > I'm trying to make a config.ttyS1 for the Cardinal modem. I keep
> > getting "unable to determine modem type" in the syslog. How do the
> > ModemMfrQueryCmd, ModemModelQueryCmd, ModemRevQueryCmd commands work?
> > I looked in the docs, and couldn't find anything.
> 
> Step back here: what operating system, what brand of Cardinal modem,
> what version of HylaFAX, and is the modem internal or external?
>

Its a Cardinal Connecta, model 3480, external 56K X2 Fax Modem,
running on Linux 1.2.8. The modem claims to support class 2.0, and
seems to be a US Robotics Sportster clone, which I am currently using
with HylaFAX in class 1. I wanted to see if the 2.0 worked where the
USR's doesn't. The Connecta seems to use the same command set as the
USR, (The I*'s formats may be different,) and the same TI DSP chip. I
tried the USR's config file for 2.0, but it gave grumblings in the
syslog.

I got the modem as an inexpensive add on for UUCP, started using it
for diald and pppd, and the firmware seemed robust. So, I thought I
would give it a test drive in class 2.0.

	John

-- 

John Conover, 631 Lamont Ct., Campbell, CA., 95008, USA.
VOX 408.370.2688, FAX 408.379.9602
john@johncon.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Sun, 9 Nov 1997 14:30:18 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: John Conover <john@johncon.johncon.com>
Cc: flexfax@sgi.com
Subject: Re: flexfax: Cardinal modem.
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Sat, 8 Nov 1997, John Conover wrote:

> Nico Garcia writes:
> > 
> > > I'm trying to make a config.ttyS1 for the Cardinal modem. I keep
> > > getting "unable to determine modem type" in the syslog. How do the
> > > ModemMfrQueryCmd, ModemModelQueryCmd, ModemRevQueryCmd commands work?
> > > I looked in the docs, and couldn't find anything.
> > 
> > Step back here: what operating system, what brand of Cardinal modem,
> > what version of HylaFAX, and is the modem internal or external?
> >
> 
> Its a Cardinal Connecta, model 3480, external 56K X2 Fax Modem,
> running on Linux 1.2.8. The modem claims to support class 2.0, and
> seems to be a US Robotics Sportster clone, which I am currently using
> with HylaFAX in class 1. I wanted to see if the 2.0 worked where the
> USR's doesn't. The Connecta seems to use the same command set as the
> USR, (The I*'s formats may be different,) and the same TI DSP chip. I
> tried the USR's config file for 2.0, but it gave grumblings in the
> syslog.

OK. I strongly suspect that the 2.0 is the problem. We've seen a lot
of reports on this group os people using modems that *claimed* to be
2.0, especially on the box, but when run only worked in Class 1 mode.

This seems especially true of the 33.6 modems, which I tried with a
USR Sportster 33.6 and a PPI 33.6 (I never did get the PPI
straightened out and had to give it back to its owner). It may
also be true of the 56k modems.

> I got the modem as an inexpensive add on for UUCP, started using it
> for diald and pppd, and the firmware seemed robust. So, I thought I
> would give it a test drive in class 2.0.

Completely reasonable approach. One note: is there any reason not to
upgrade your Linux to a more recent, 2.0.30 kernel? There are a *lot*
of patches, drivers, and security fixes in the more recent kernels
that aren't available in the older ones.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNGYPQz/+ItycgIJRAQFOEAQAhJQO2hWHHd8MLV1Szm3q9/xOKGqcaGPo
Ef4I1Tvq+Xmpz4iWF13im6SvGb7eL3ON+d4wUs104tgNhN4Fzwhl2pafjoD5C+qK
in2oCdl47feTzOHFeeM1J0XKK0ag5FhAYzvtpE8rndvKlALnTrDOA1apVmrXzKLI
taIGB6oesCY=
=73XP
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
X-Authentication-Warning: tui.pinnacle.co.nz: jonc owned process doing -bs
Date: Mon, 10 Nov 1997 09:01:53 +1300 (NZDT)
From: Jonathan Chen <jonc@pinnacle.co.nz>
To: Ancil Bethelmy <cowpoke@wave.cape.net>
cc: Hylafax Mailing list <flexfax@sgi.com>
Subject: Re: flexfax: What device name for com1 on dec alpha
Sender: owner-flexfax@celestial.com

On Fri, 7 Nov 1997, Ancil Bethelmy wrote:

> Thanks. This is a major help.
> 
> I just got of the phone with Dec support and they indicated that I need to
> run uucpsetup to get the modem devices configured. He also indicated that I
> needed to use /dev/ttyd0 as the modem device. Do you know anything about
> that?

That's unusual. I've got HylaFAX running at one of our client sites,
and I don't recall having to run uucpsetup to configure anything at
all. /dev/tty00 works fine.

What sort of problems are your having? Give version numbers of
everything you're running (to the list - note Cc:).

Cheers.
--
Jonathan Chen <jonc@pinnacle.co.nz>                 Once is dumb luck.
                                                 Twice is coincidence.
             Three times and Somebody Is Trying To Tell You Something.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Sun, 9 Nov 1997 15:21:46 -0700 (MST)
From: David Hughes <davidh@csn.net>
Subject: flexfax: modem on /dev/cua1 appears wedged (fwd)
To: flexfax@sgi.com
Sender: owner-flexfax@celestial.com


As a note, I have been searching the HylaFAX archives and seen mention of 
the following, but haven't been able to resolve my problem below yet.

.......


I am attempting to install HylaFAX on RedHat4.2/Intel and have received the
following error email to root.

The same info is issued in /var/log/messages.  I disable the inittab line
for this so that /dev/cua1 is not locked, and then run minicom and chat
with the modem just fine, issuing modem or faxmodem commands with no
problem.  But each time I restart the system, it craps out on this error. 
I have tried two separate modems (they are PCMCIA cards which run fine
with everything else). 

I have attempted to rerun faxaddmodem and every time it senses the modem
and spits back that it is talking back and forth fine.  It configures the
modem and chatters back and forth with no problem, telling me the modem
type, class type recommendation, etc.  But then on system restart, it says
the modem is 'wedged'. 

Help anyone?  Anyone successfully installed HylaFAX onto a laptop using
PCMCIA modems?  One is a Megahertz 14.4 FaxModem and the second is a US 
Robotics Megahertz 56K X2.  Both have run fine in all other capacities.

Error message listed below. . . 

As a side note, I just subscribed to the flexfax list and am impressed 
with all the email that has already come across.  I hope someone will be 
able to help with this.  

thanks, david





>Date: Sun, 9 Nov 1997 14:02:56 -0700
>From: The HylaFAX Receive Agent <fax@david.santuario.com>
>Subject: modem on /dev/cua1 appears wedged

>The HylaFAX software thinks that there is a problem with the modem
>on device /dev/cua1 that needs attention; repeated attempts to
>initialize the modem have failed.

>Consult the server trace logs for more information on what is happening.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 7 Nov 1997 11:46:21 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Matthias Apitz <Matthias.Apitz@SOFTCON.de>
Cc: Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: sharing a modem with pppd
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Fri, 7 Nov 1997 Matthias.Apitz@SOFTCON.de wrote:

> The FAXSERVER environment variable controls which host and port
> the clients (e.g. faxstat(1) or sendfax(1)) should contact
> for the "hylafax" service of hfaxd(1M). This variable has
> nothing to do with the running hfaxd(1M) and if I understand
> the problem report of Luiz Otavio right the problem is that
> a client *can* reach the service at the host, the hfaxd(1M)
> gets the connection, forks and aborts with SIG 11 in
> case the PPP-link isn't up. This sounds like related to a name
> resolving problem but is unrelated to FAXSERVER setting.

Because FAXSERVER controls the host, and DNS under PPP's control looks
up the hostname's IP address, selecting a FAXSERVER other than the
localhost default can screw things up if the hostname *changes* when
he drops the PPP connection. I've seen this sort of PPP setup,
especially for dynamic IP address allocation. In that case, it's
important to make sure the FAXSERVER says "localhost" or is left
blank.

> At the moment I don't know which OS type Luiz Otavio is
> running.... ah, it's Linux (just checked my HylaFAX mail
> folders). Luiz Otavio, please run a strace(1) command
> with the -f flag (to monitor also forked childs) against
> the running hfaxd(1M), start this before you fire up the
> client (or just use "telnet localhost hylafax") and
> send me the output of the strace(1). I'll also send
> you a patch (directly) which deals with this problem in pl2
> (but without the strace I'm not sure if you hit the same
> situation).

Good stuff. Glad you're on it, this is deeper than I was looking...

> I also managed to compile HylaFAX at home on the S.u.S.E. Linux
> box of my son (normaly I run SVR4.2) and can now answer
> Linux related questions better than before.

Cool. I'd be glad to give the current pl2 version a test under SunOS
on a spare server I have set up, whenever you like.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNGNF0T/+ItycgIJRAQE99gP+P6w0geQ+D0Sp9IzAuInze7RMTXQJ9Lf5
fcCpwsCA9/GYA8eGXzZR8wo/83uYo2644Y5Clo7o3GSFGj5SiNgjXRfZjfpA92vL
09iIRlqdwJmhSMVbCzntE5MEgwqPPN3Hc5QhNO07lIYaP8aak4wuLlP6tog+phZd
tnSIWkUkpdI=
=GdYC
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 7 Nov 1997 11:50:59 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Alessandro Coppelli <coppelli@sun3.dsea.unipi.it>
Cc: flexfax@sgi.com
Subject: Re: flexfax: From tiff to ps and then to MSWord
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Fri, 7 Nov 1997, Alessandro Coppelli wrote:

>   I would to want to get one received fax ( format tiff ) and read it then with
>   
>   MSWord 97 . How I can to do ?.
>   
>   I have one server fax ( hylafax in Linux 2.0.30 box ) .
>   
>   I must use tiff2ps ?

Use fax2ps. 

>   and then how I can read it in MSWord 97 ?

You won't get words: the best you will get will be an *image* of the
fax page. Turning it into words would require a decent On-Line
Character Recognition program, such as what comes with the Paperport
scanners.

If they want to send you a document as words or a document you can
edit, they should send you email....

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNGNG7D/+ItycgIJRAQFdoQP7ByUYc8ZxhDM82LzDHAiyhAZDVwy0nfp7
iOz/g/RbT41ZXxobMLJ6NLT3MpQFYagAkaDA2iwRvfDo2b5rmj3vp0raEJAMiKdL
CnBX5vk5ZNukKKyoUMS1Ce1ajpqBrgEOY89Ryp2KobDfcA1z8k+o+hdJPIiKDkJS
rLtS0XIId70=
=i4tR
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Leslie Mikesell <les@Mcs.Net>
Subject: Re: flexfax: How best use multiple modems?
To: Matthias.Apitz@SOFTCON.de
Date: Fri, 7 Nov 1997 11:24:07 -0600 (CST)
Cc: spine001@gold.tc.umn.edu, flexfax@sgi.com
Sender: owner-flexfax@celestial.com

>     
>     > Modems connected to network servers will not work reliable with
>     > HylaFAX because you can't control the time over a TCP/IP link
>     > to match the T.30 timer demands.
>     > 
>     > 
>     Can we use modems that support class 2 Rockwell type command set over 
>     the TCP/IP network instead of connecting them to the serial devices? 
>     Would they handle the T.30 timer demands internally?
> 
> The modem will handle the timer correctly (if it does handle
> them correctly et all) but you can't make sure to pump the
> data or AT-cmds to the modem (or read answers from the modem)
> in the timer intervalls.

TCP doesn't guarantee the timing but an ethernet has to be pretty badly
loaded to impose significant delays handling something like a 19200 baud
serial stream.  I'd expect this to be possible if both ends do something
reasonable with the delay trying to accumulate characters into packets.
You can control this with TCP_NODELAY on the unix side, but I don't
know if this is tunable on the terminal serve side.  Also, with the
current version of HylaFax you have to go through another program
that maps a pty to the socket which is probably not very efficient.
It might help if HylaFax were able to open a socket directly to
the modem.  

 Les Mikesell
   les@mcs.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Comments: Authenticated sender is <spine001@gold.tc.umn.edu>
From: Julio C Spinelli <spine001@gold.tc.umn.edu>
To: Leslie Mikesell <les@Mcs.Net>
Date: Sat, 8 Nov 1997 08:30:44 +0000
Subject: Re: flexfax: How best use multiple modems?
CC: Julio C Spinelli <spine001@gold.tc.umn.edu>, flexfax@sgi.com
Sender: owner-flexfax@celestial.com

Thanks for the clarification, but could you illuminate us and tell us 
what this program is that can allow you to map the pty to the 
socket?

Thanks in advance,

Julio
> From:          Leslie Mikesell <les@Mcs.Net>
> Subject:       Re: flexfax: How best use multiple modems?
> To:            Matthias.Apitz@SOFTCON.de
> Date:          Fri, 7 Nov 1997 11:24:07 -0600 (CST)
> Cc:            Julio C Spinelli <spine001@gold.tc.umn.edu>,
>                flexfax@sgi.com

> >     
> >     > Modems connected to network servers will not work reliable with
> >     > HylaFAX because you can't control the time over a TCP/IP link
> >     > to match the T.30 timer demands.
> >     > 
> >     > 
> >     Can we use modems that support class 2 Rockwell type command set over 
> >     the TCP/IP network instead of connecting them to the serial devices? 
> >     Would they handle the T.30 timer demands internally?
> > 
> > The modem will handle the timer correctly (if it does handle
> > them correctly et all) but you can't make sure to pump the
> > data or AT-cmds to the modem (or read answers from the modem)
> > in the timer intervalls.
> 
> TCP doesn't guarantee the timing but an ethernet has to be pretty badly
> loaded to impose significant delays handling something like a 19200 baud
> serial stream.  I'd expect this to be possible if both ends do something
> reasonable with the delay trying to accumulate characters into packets.
> You can control this with TCP_NODELAY on the unix side, but I don't
> know if this is tunable on the terminal serve side.  Also, with the
> current version of HylaFax you have to go through another program
> that maps a pty to the socket which is probably not very efficient.
> It might help if HylaFax were able to open a socket directly to
> the modem.  
> 
>  Les Mikesell
>    les@mcs.com
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Leslie Mikesell <les@Mcs.Net>
Subject: Re: flexfax: How best use multiple modems?
To: spine001@gold.tc.umn.edu (Julio C Spinelli)
Date: Sat, 8 Nov 1997 11:40:03 -0600 (CST)
Cc: les@Mcs.Net, spine001@gold.tc.umn.edu, flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> 
> Thanks for the clarification, but could you illuminate us and tell us 
> what this program is that can allow you to map the pty to the 
> socket?

Poke around the ftp areas of the vendor of your terminal server.  There
is one at ftp://ftp.livingston.com/pub/livingston/contrib/netty.c
but it may be portmaster-specific in the way it treats the modems
as a pool.  I think cisco used to have a similar program called tcpconn
but they have rearranged things and I don't see it now.  Anyway,
handling ptys is somewhat os-specific and the socket side is fairly
simple.  If anyone needs this I think they'd be better off modifying
Hylafax to bypass the pty and talk directly to a socket thus avoiding
the extra latency of a program in the middle.

I've always thought it would be nice if Hylafax also provided the
reverse service as well, emulating the portmaster function where
you connect to a tcp port and get an available modem from the
pool.  Faxq already knows the states of all the modems so it
could pick one easily and launch a program that simply copies
between socket/device.  Then you could use the shareware comt
program to let Windows machines on the net share the same modems
used by Hylafax.

  Les Mikesell
    les@mcs.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
cc: raoul@cirl.meei.harvard.edu (Nico Garcia), lists@comconnect.com,
        flexfax@sgi.com
Subject: Re: flexfax: How best use multiple modems? 
Date: Sun, 09 Nov 1997 15:48:15 +0000
From: "Mr. Arlington Hewes" <tpcadmin@info.tpc.int>
Sender: owner-flexfax@celestial.com

>>>>> On Fri, 7 Nov 1997, "MA" == Matthias.Apitz@SOFTCON.de wrote:

  MA> Modems connected to network servers will not work reliable with
  MA> HylaFAX because you can't control the time over a TCP/IP link to match
  MA> the T.30 timer demands.

On a lossy network with a lot of packet retransmits, I admit that we'd be
beating a dead horse here. But I can guarantee, as can most/many who have
Ascend boxes laying around, that the MAX would be on the same carefully
partitioned subnet as the HylaFAX box.


les@Mcs.Net said:
> Also, with the current version of HylaFax you have to go through
> another program that maps a pty to the socket which is probably not
> very efficient. It might help if HylaFax were able to open a socket
> directly to the modem.

I'm very interested in adding this functionality to HylaFAX, and I know
Ascends already have some hooks for it and are developing others. I might
even be able to pay some willing code-warrior to develop it, and then fold
it back into HylaFAX. Anyone interested?

-DPN

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 07 Nov 1997 19:26:53 +0100
From: Ramon Arxer <arxer@bcn.servicom.es>
Reply-To: arxer@bcn.servicom.es
Organization: http://yi.com/home/ArxerRamon/
To: flexfax list <flexfax@sgi.com>
Subject: flexfax: sendfax doesn't work?
Sender: owner-flexfax@celestial.com

Hello!

I'm trying to install HylaFAX again (this is the second time), the first
time (at home) it worked very well, and now at the school it seems that
some problems appear.
I'm installing HylaFAX 4.0pl1 with the Dirks-faxmailer patch on a
Pentium based server with Linux.

When I send a fax the faxmailer script executes the following sentence:

/usr/local/bin/faxmail -n | /usr/local/bin/sendfax -f 'root <root> (+34
3 401-68-01)' -h 'gea04' -n -m -d '@4114343' -D 2>&1
1>/tmp/.faxmailer.166

then the sendfax program stops. With the ps -x command you can see:

165  p2 S     0:00 send-mail -i fax.4114343 
166  ?  S     0:00 perl /usr/local/bin/faxmailer fax 4114343 
168  ?  S     0:00 sh -c /usr/local/bin/faxmail -n |
/usr/local/bin/sendfax -f
170  ?  S     0:00 /usr/local/bin/sendfax -f root <root> (+34 3
401-68-01) -h

And in the log file all seems stopped also:

Nov  7 19:52:15 gea04 FaxQueuer[124]: HylaFAX (tm) Version 4.0pl1
Nov  7 19:52:15 gea04 FaxQueuer[124]: Copyright (c) 1990-1996 Sam
Leffler
Nov  7 19:52:15 gea04 FaxQueuer[124]: Copyright (c) 1991-1996 Silicon
Graphics, Inc.
Nov  7 19:52:18 gea04 HylaFAX[134]: HylaFAX INET Protocol Server:
restarted.
Nov  7 19:52:18 gea04 HylaFAX[134]: HylaFAX Old Protocol Server:
restarted.
Nov  7 19:52:18 gea04 HylaFAX[134]: HylaFAX SNPP Protocol Server:
restarted.
Nov  7 19:52:21 gea04 FaxGetty[141]: OPEN /dev/cua1
Nov  7 19:52:21 gea04 FaxGetty[141]: MODEM set DTR OFF
Nov  7 19:54:42 gea04 sendmail[164]: TAA00164: from=root, size=43,
class=0, pri=30043, nrcpts=1,
msgid=<199711071854.TAA00164@gea04.pangea.org>, relay=root@localhost
Nov  7 19:54:43 gea04 HylaFAX[171]: Filesystem has SysV-style file
creation semantics.

Somebody has an idea?

Thank you in advance.
-- 

Ramon Arxer i Galabert


e-mail:
	arxer@bcn.servicom.es
	ramon25@casal.upc.es
WWW:
	http://yi.com/home/ArxerRamon/
	http://casal.upc.es/~ramon25/

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: sendfax doesn't work?
To: arxer@bcn.servicom.es
Date: Sat, 8 Nov 1997 08:45:34 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Ramon Arxer wrote:
    
    Hello!
    
    I'm trying to install HylaFAX again (this is the second time), the first
    time (at home) it worked very well, and now at the school it seems that
    some problems appear.
    I'm installing HylaFAX 4.0pl1 with the Dirks-faxmailer patch on a
    Pentium based server with Linux.
    
    When I send a fax the faxmailer script executes the following sentence:
    
    /usr/local/bin/faxmail -n | /usr/local/bin/sendfax -f 'root <root> (+34
    3 401-68-01)' -h 'gea04' -n -m -d '@4114343' -D 2>&1
    1>/tmp/.faxmailer.166
    
    then the sendfax program stops. With the ps -x command you can see:
    
    165  p2 S     0:00 send-mail -i fax.4114343 
    166  ?  S     0:00 perl /usr/local/bin/faxmailer fax 4114343 
    168  ?  S     0:00 sh -c /usr/local/bin/faxmail -n |
    /usr/local/bin/sendfax -f
    170  ?  S     0:00 /usr/local/bin/sendfax -f root <root> (+34 3
    401-68-01) -h
    
    And in the log file all seems stopped also:
    
    Nov  7 19:52:15 gea04 FaxQueuer[124]: HylaFAX (tm) Version 4.0pl1
    Nov  7 19:52:15 gea04 FaxQueuer[124]: Copyright (c) 1990-1996 Sam
    Leffler
    Nov  7 19:52:15 gea04 FaxQueuer[124]: Copyright (c) 1991-1996 Silicon
    Graphics, Inc.
    Nov  7 19:52:18 gea04 HylaFAX[134]: HylaFAX INET Protocol Server:
    restarted.
    Nov  7 19:52:18 gea04 HylaFAX[134]: HylaFAX Old Protocol Server:
    restarted.
    Nov  7 19:52:18 gea04 HylaFAX[134]: HylaFAX SNPP Protocol Server:
    restarted.
    Nov  7 19:52:21 gea04 FaxGetty[141]: OPEN /dev/cua1
    Nov  7 19:52:21 gea04 FaxGetty[141]: MODEM set DTR OFF
    Nov  7 19:54:42 gea04 sendmail[164]: TAA00164: from=root, size=43,
    class=0, pri=30043, nrcpts=1,
    msgid=<199711071854.TAA00164@gea04.pangea.org>, relay=root@localhost
    Nov  7 19:54:43 gea04 HylaFAX[171]: Filesystem has SysV-style file
    creation semantics.
    
    Somebody has an idea?

Try running telnet on the same host where the sendfax(1) is running

$ telnet gea04 4559

and see what the hfaxd(1M) is doing for you. You may also enable
the trace of the client/server protocol in the hfaxd.conf file.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: flexfax: faxsetup on Linux (bug in moduser?)
To: flexfax@sgi.com
Date: Sun, 9 Nov 1997 16:47:20 +0100 (MEZ)
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com


I'm checking the faxsetup(1M) on Linux (to detect any problems
and see if it works reliable). For some unknown reason my Linux
box (S.u.S.E.) is shipped with a user "fax" in /etc/passwd and
HylaFAX's faxsetup(1M) wants to modify this ("fax" must have
the same uid as "uucp"). But it seems that there is a bug
in the usermod(1M) command. It complains about the existing
(new) HOME of "fax" with exit code 12 but the change is done
in the /etc/passwd file. A workaround could be running the command
twice 'cuze it exits with 0 for the 2nd change which has
nothing to do anymore. Any comments from the Linux gurus?

# uname -a
Linux map 2.0.30 #9 Wed Nov 5 18:41:26 MET 1997 i586 unknown
# grep fax /etc/passwd
fax:x:23:14:Facsimile Agent:/var/spool/fax:/bin/bash
# usermod -d /usr/local/lib/HylaFAX/spool -m -u 10 -o -g 14 fax
usermod: directory /usr/local/lib/HylaFAX/spool exists
# echo $?
12
# grep fax /etc/passwd
fax:x:10:14:Facsimile Agent:/usr/local/lib/HylaFAX/spool:/bin/bash
# 
# usermod -d /usr/local/lib/HylaFAX/spool -m -u 10 -o -g 14 fax
# echo $?
0
#

BTW: The usermod(1M) wants the flag "-m" after the "-d home"
while the original faxsetup script in pl1 has the "-m" as the
first parameter. This can easy be fixed in the script faxsetup.linux
(or should be fixed in faxsetup at all).

All other things in faxsetup(1M) work without any kind of problems.
The faxq(1M) is up and the hfaxd(1M) starts on connects via
the inetd(1M):

thias:/home/guru $ FAXSERVER=map:4559 export FAXSERVER
thias:/home/guru $ faxstat -v
HylaFAX scheduler on map.muc.de: Running
Trying map (222.222.222.223) at port 4559...
Connected to map.
220 map.muc.de server (HylaFAX (tm) Version 4.0pl2beta) ready.
-> USER guru
230 User guru logged in.
-> PORT 222,222,222,222,5,152
200 PORT command successful.
-> LIST status
150 Opening new data connection for "status".
226 Transfer complete.
thias:/home/guru $ 


        matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Sun, 9 Nov 1997 14:40:58 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Matthias Apitz <Matthias.Apitz@SOFTCON.de>
Cc: flexfax@sgi.com
Subject: Re: flexfax: faxsetup on Linux (bug in moduser?)
Sender: owner-flexfax@celestial.com

On Sun, 9 Nov 1997 Matthias.Apitz@SOFTCON.de wrote:

> I'm checking the faxsetup(1M) on Linux (to detect any problems
> and see if it works reliable). For some unknown reason my Linux
> box (S.u.S.E.) is shipped with a user "fax" in /etc/passwd and
> HylaFAX's faxsetup(1M) wants to modify this ("fax" must have
> the same uid as "uucp"). But it seems that there is a bug
> in the usermod(1M) command. It complains about the existing
> (new) HOME of "fax" with exit code 12 but the change is done
> in the /etc/passwd file. A workaround could be running the command
> twice 'cuze it exits with 0 for the 2nd change which has
> nothing to do anymore. Any comments from the Linux gurus?

I think the problem is slightly different flavors of Linux have
slightly different flavors of the adduser or usermod or whatever
command. I think this is a completely reasonable workaround.  Some use
passmgmt, some use useradd, some use adduser, etc.

> BTW: The usermod(1M) wants the flag "-m" after the "-d home"
> while the original faxsetup script in pl1 has the "-m" as the
> first parameter. This can easy be fixed in the script faxsetup.linux
> (or should be fixed in faxsetup at all).

faxsetup.linux would be the place to do it, I think.
> All other things in faxsetup(1M) work without any kind of problems.
> The faxq(1M) is up and the hfaxd(1M) starts on connects via
> the inetd(1M):

I thought most of us were using the boot-time script to start faxq and
hfaxd as background daemons. Is the inetd.conf technique now
considered reliable enough for general use on low-volume sites?

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: faxsetup on Linux (bug in moduser?)
To: raoul@cirl.meei.harvard.edu (Nico Garcia)
Date: Mon, 10 Nov 1997 08:11:40 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Nico Garcia wrote:
    
    On Sun, 9 Nov 1997 Matthias.Apitz@SOFTCON.de wrote:
    
    > I'm checking the faxsetup(1M) on Linux (to detect any problems
    > and see if it works reliable). For some unknown reason my Linux
    > box (S.u.S.E.) is shipped with a user "fax" in /etc/passwd and
    > HylaFAX's faxsetup(1M) wants to modify this ("fax" must have
    > the same uid as "uucp"). But it seems that there is a bug
    > in the usermod(1M) command. It complains about the existing
    > (new) HOME of "fax" with exit code 12 but the change is done
    > in the /etc/passwd file. A workaround could be running the command
    > twice 'cuze it exits with 0 for the 2nd change which has
    > nothing to do anymore. Any comments from the Linux gurus?
    
    I think the problem is slightly different flavors of Linux have
    slightly different flavors of the adduser or usermod or whatever
    command. I think this is a completely reasonable workaround.  Some use
    passmgmt, some use useradd, some use adduser, etc.

The HylaFAX software can't work corretly if the user "fax"
in the passwd(4) has a different uid than "uucp". If there
is such wrong entry the faxsetup(1M) script should be able to
fix this. I'll let this to the Linux folks to clear what's to
do to have a working usermod(1M) command in Linux.
For the moment I'll call it twice in faxsetup.linux.
    
    > BTW: The usermod(1M) wants the flag "-m" after the "-d home"
    > while the original faxsetup script in pl1 has the "-m" as the
    > first parameter. This can easy be fixed in the script faxsetup.linux
    > (or should be fixed in faxsetup at all).
    
    faxsetup.linux would be the place to do it, I think.

Not sure. The SVR4 usermod(1M) also wants the "-m" flag after
the "-d home".

    > All other things in faxsetup(1M) work without any kind of problems.
    > The faxq(1M) is up and the hfaxd(1M) starts on connects via
    > the inetd(1M):
    
    I thought most of us were using the boot-time script to start faxq and
    hfaxd as background daemons. Is the inetd.conf technique now
    considered reliable enough for general use on low-volume sites?

Yes, that's the message of my posting. Running hfaxd(1M) from
inetd.conf(4) should work. If it does not, someone should fix
it on the OS where it causes problems. The line in Linux's
inetd.conf(4) (added by faxsetup(1M)) is:

hylafax stream  tcp     nowait  fax     /usr/local/lib/HylaFAX/sbin/hfaxd      hfaxd -I

and faxsetup(1M) sent also SIG_HUP to the inetd(1M) to let it
re-read the file inetd.conf(1M). It's working fine and there is no
reason not to use faxsetup(1M) on Linux.

On my production system I'm running hfaxd(1M) from a boot-time
script in /etc/rc2.d/  Both techniques should work.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 10 Nov 1997 14:12:36 +0100 (MET)
From: Carsten Hoeger <choeger@suse.de>
To: Matthias Apitz <Matthias.Apitz@SOFTCON.de>
cc: flexfax@sgi.com
Subject: Re: flexfax: faxsetup on Linux (bug in moduser?)
Sender: owner-flexfax@celestial.com

On Sun, 9 Nov 1997 Matthias.Apitz@SOFTCON.de wrote:

> I'm checking the faxsetup(1M) on Linux (to detect any problems
> and see if it works reliable). For some unknown reason my Linux
> box (S.u.S.E.) is shipped with a user "fax" in /etc/passwd and

The reason is, in my opinion the package mgetty+sendfax from Gert Doering.
This package needs this user, too.

> HylaFAX's faxsetup(1M) wants to modify this ("fax" must have
> the same uid as "uucp"). But it seems that there is a bug

By this reason, I made some very raw changes in the faxsetup-script from
HylaFAX, that comes with S.u.S.E. Linux 5.0.0#2 and higher. After these
changes, the question to modify /etc/passwd disappears -that's not a very
good solution, but I had to find a quick solution-.

Another point is, that the uid of uucp and fax has to be the same. That's
not a very good solution. We changed this and put uucp and fax in the same
group. That works...

> in the usermod(1M) command. It complains about the existing
> (new) HOME of "fax" with exit code 12 but the change is done
> in the /etc/passwd file. A workaround could be running the command
> twice 'cuze it exits with 0 for the 2nd change which has
> nothing to do anymore. Any comments from the Linux gurus?

...see above...
 
> # uname -a
> Linux map 2.0.30 #9 Wed Nov 5 18:41:26 MET 1997 i586 unknown
> # grep fax /etc/passwd
> fax:x:23:14:Facsimile Agent:/var/spool/fax:/bin/bash
> # usermod -d /usr/local/lib/HylaFAX/spool -m -u 10 -o -g 14 fax
> usermod: directory /usr/local/lib/HylaFAX/spool exists
> # echo $?
> 12
> # grep fax /etc/passwd
> fax:x:10:14:Facsimile Agent:/usr/local/lib/HylaFAX/spool:/bin/bash
> # 
> # usermod -d /usr/local/lib/HylaFAX/spool -m -u 10 -o -g 14 fax
> # echo $?
> 0
> #
> 
> BTW: The usermod(1M) wants the flag "-m" after the "-d home"
> while the original faxsetup script in pl1 has the "-m" as the
> first parameter. This can easy be fixed in the script faxsetup.linux
> (or should be fixed in faxsetup at all).

I'm very pleased about this, but why uid(uucp) == uid(fax)?

> All other things in faxsetup(1M) work without any kind of problems.
> The faxq(1M) is up and the hfaxd(1M) starts on connects via
> the inetd(1M):
> 
> thias:/home/guru $ FAXSERVER=map:4559 export FAXSERVER
> thias:/home/guru $ faxstat -v
> HylaFAX scheduler on map.muc.de: Running
> Trying map (222.222.222.223) at port 4559...
> Connected to map.
> 220 map.muc.de server (HylaFAX (tm) Version 4.0pl2beta) ready.
> -> USER guru
> 230 User guru logged in.
> -> PORT 222,222,222,222,5,152
> 200 PORT command successful.
> -> LIST status
> 150 Opening new data connection for "status".
> 226 Transfer complete.
> thias:/home/guru $ 

mfG,
	Carsten Hoeger
------
Carsten Hoeger  - S.u.S.E. GmbH -  Gebhardtstr. 2  -  90762 Fuerth  -  Germany
fax +49-911-3206727      web http://www.suse.de        mailbox +49-911-3247122
"Ein Buch ist eine Wolke, auf der man keinen Husten bekommt." (Max Goldt)

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 10 Nov 1997 09:12:19 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
Cc: Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: faxsetup on Linux (bug in moduser?)
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Mon, 10 Nov 1997, Carsten Hoeger wrote:

> Another point is, that the uid of uucp and fax has to be the same. That's
> not a very good solution. We changed this and put uucp and fax in the same
> group. That works...

Ummm. This is not ideal. Dial-up lines and lock files are normally
owned by the calling process, UUCP or HylaFAX. Using the same group
for the uucp/fax users should only work if the lock files are left
with group write permission, and the serial lines left owned by that
user. I'd want to think about that before doing it.

It also involves mucking with the group files as well as the passwd
file...

> > BTW: The usermod(1M) wants the flag "-m" after the "-d home"
> > while the original faxsetup script in pl1 has the "-m" as the
> > first parameter. This can easy be fixed in the script faxsetup.linux
> > (or should be fixed in faxsetup at all).
> 
> I'm very pleased about this, but why uid(uucp) == uid(fax)?

See above.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNGcWNz/+ItycgIJRAQGrWwP9EeRwMSpnE7RZRXOBNEhHo2BSVDqu9YM6
lXv+wjqx3y8vphAFtX4HCu4F38dj/T5kElMo4fTO4QRKjBKdVlUoCSVgL8uNkdCi
5Tt5dmapzoyeYvmozWEVqcLgdxhLAb/tDWc6FRnSx9GXB1Z3CNBsj3jCBwJei1JG
+j4LUi5E77g=
=yBrY
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: faxsetup on Linux (bug in moduser?)
To: choeger@suse.de (Carsten Hoeger)
Date: Mon, 10 Nov 1997 15:58:39 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Carsten Hoeger wrote:
    
    On Sun, 9 Nov 1997 Matthias.Apitz@SOFTCON.de wrote:
    
    > I'm checking the faxsetup(1M) on Linux (to detect any problems
    > and see if it works reliable). For some unknown reason my Linux
    > box (S.u.S.E.) is shipped with a user "fax" in /etc/passwd and
    
    The reason is, in my opinion the package mgetty+sendfax from Gert Doering.
    This package needs this user, too.
    
    > HylaFAX's faxsetup(1M) wants to modify this ("fax" must have
    > the same uid as "uucp"). But it seems that there is a bug
    
    By this reason, I made some very raw changes in the faxsetup-script from
    HylaFAX, that comes with S.u.S.E. Linux 5.0.0#2 and higher. After these
    changes, the question to modify /etc/passwd disappears -that's not a very
    good solution, but I had to find a quick solution-.

    Another point is, that the uid of uucp and fax has to be the same. That's
    not a very good solution. We changed this and put uucp and fax in the same
    group. That works...

The reason for uid(uucp) == uid(fax) are the lock files of UUCP.
The normal directory for UUCP-locking is /var/spool/locks,
owned by "uucp" with permission 0755. I just checked another
Linux box in my company (also a S.u.S.E. distribution)
but there seems to be a bug (don't know if this system still
matches the CD-ROM):

$ uname -a
Linux wilson 2.0.14 #5 Thu Aug 22 17:01:26 MET DST 1996 i586
$ ls -ld /var/spool/locks
lrwxrwxrwx   1 root     root            4 Mar 17  1995 /var/spool/locks -> lock
$ cd /var/spool/locks
bash: /var/spool/locks: No such file or directory
$

If you set "fax" and "uucp" into the same group (e.g. "uucp")
you must (at least) also change the rights of this directory.

If mgetty+sendfax can live with any uid for "fax" why not
setting it to the same as "uucp" per default in S.u.S.E.?

Do you have any information why usermod(1M) is exiting with 12
in case of existing HOME dir? This is the real problem at all.
On my SVR4.2 Onsite UNIX usermod(1M) works like that:

# grep fax /etc/passwd
fax:x:5:5:Facsimile Agent:/var/spool/fax:/usr/bin/sh
# usermod -d /tmp fax
# echo $?
0
# grep fax /etc/passwd
fax:x:5:5:Facsimile Agent:/tmp:/usr/bin/sh
# 

Another workaround could be not to change the HOME for "fax".
HylaFAX doesn't care about fax's $HOME.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 10 Nov 1997 16:58:31 +0100 (MET)
From: Carsten Hoeger <choeger@suse.de>
To: Matthias.Apitz@SOFTCON.de, Matthias Apitz <guru@sisis.de>
cc: flexfax@sgi.com
Subject: Re: flexfax: faxsetup on Linux (bug in moduser?)
Sender: owner-flexfax@celestial.com

On Mon, 10 Nov 1997 Matthias.Apitz@SOFTCON.de wrote:

> Carsten Hoeger wrote:
>     
>     On Sun, 9 Nov 1997 Matthias.Apitz@SOFTCON.de wrote:
>     
>     > I'm checking the faxsetup(1M) on Linux (to detect any problems
>     > and see if it works reliable). For some unknown reason my Linux
>     > box (S.u.S.E.) is shipped with a user "fax" in /etc/passwd and
>     
>     The reason is, in my opinion the package mgetty+sendfax from Gert Doering.
>     This package needs this user, too.
>     
>     > HylaFAX's faxsetup(1M) wants to modify this ("fax" must have
>     > the same uid as "uucp"). But it seems that there is a bug
>     
>     By this reason, I made some very raw changes in the faxsetup-script from
>     HylaFAX, that comes with S.u.S.E. Linux 5.0.0#2 and higher. After these
>     changes, the question to modify /etc/passwd disappears -that's not a very
>     good solution, but I had to find a quick solution-.
> 
>     Another point is, that the uid of uucp and fax has to be the same. That's
>     not a very good solution. We changed this and put uucp and fax in the same
>     group. That works...
> 
> The reason for uid(uucp) == uid(fax) are the lock files of UUCP.
> The normal directory for UUCP-locking is /var/spool/locks,
> owned by "uucp" with permission 0755. I just checked another
> Linux box in my company (also a S.u.S.E. distribution)
> but there seems to be a bug (don't know if this system still
> matches the CD-ROM):

The uucp-locking directory is /var/lock in our system...
 
> $ uname -a
> Linux wilson 2.0.14 #5 Thu Aug 22 17:01:26 MET DST 1996 i586
> $ ls -ld /var/spool/locks
> lrwxrwxrwx   1 root     root            4 Mar 17  1995 /var/spool/locks -> lock
> $ cd /var/spool/locks
> bash: /var/spool/locks: No such file or directory
> $
> 
> If you set "fax" and "uucp" into the same group (e.g. "uucp")
> you must (at least) also change the rights of this directory.
> 
> If mgetty+sendfax can live with any uid for "fax" why not
> setting it to the same as "uucp" per default in S.u.S.E.?

Some guys here weren't very happy with this solution, so I tried to give
fax another uid, changed the faxaddmodem to give the FIFO.xxx file
uucp.uucp as owner and all seems to work -all I have tested-. I use the
faxserver at home, at our company and several other people use it, too.
 
> Do you have any information why usermod(1M) is exiting with 12
> in case of existing HOME dir? This is the real problem at all.
> On my SVR4.2 Onsite UNIX usermod(1M) works like that:

I'll try to find it out...
 
> # grep fax /etc/passwd
> fax:x:5:5:Facsimile Agent:/var/spool/fax:/usr/bin/sh
> # usermod -d /tmp fax
> # echo $?
> 0
> # grep fax /etc/passwd
> fax:x:5:5:Facsimile Agent:/tmp:/usr/bin/sh
> # 
> 
> Another workaround could be not to change the HOME for "fax".
> HylaFAX doesn't care about fax's $HOME.
> 
> 	matthias
> 

mfG,
	Carsten Hoeger
------
Carsten Hoeger  - S.u.S.E. GmbH -  Gebhardtstr. 2  -  90762 Fuerth  -  Germany
fax +49-911-3206727      web http://www.suse.de        mailbox +49-911-3247122
"Ein Buch ist eine Wolke, auf der man keinen Husten bekommt." (Max Goldt)

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 10 Nov 1997 18:32:59 +0100 (MET)
From: Carsten Hoeger <choeger@suse.de>
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
cc: flexfax@sgi.com
Subject: Re: flexfax: faxsetup on Linux (bug in moduser?)
Sender: owner-flexfax@celestial.com

On Mon, 10 Nov 1997, Nico Garcia wrote:

> On Mon, 10 Nov 1997, Carsten Hoeger wrote:
> 
> > Another point is, that the uid of uucp and fax has to be the same. That's
> > not a very good solution. We changed this and put uucp and fax in the same
> > group. That works...
> 
> Ummm. This is not ideal. Dial-up lines and lock files are normally
> owned by the calling process, UUCP or HylaFAX. Using the same group
> for the uucp/fax users should only work if the lock files are left
> with group write permission, and the serial lines left owned by that
> user. I'd want to think about that before doing it.

Why not rw-permissions for the 'dialin' groups lock-files?
Do you think that this may be a security hole?


mfG,
	Carsten Hoeger
------
Carsten Hoeger  - S.u.S.E. GmbH -  Gebhardtstr. 2  -  90762 Fuerth  -  Germany
fax +49-911-3206727      web http://www.suse.de        mailbox +49-911-3247122
"Ein Buch ist eine Wolke, auf der man keinen Husten bekommt." (Max Goldt)

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: faxsetup on Linux (bug in moduser?)
To: choeger@suse.de (Carsten Hoeger)
Date: Mon, 10 Nov 1997 18:39:30 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Carsten Hoeger wrote:
    
    >     Another point is, that the uid of uucp and fax has to be the same. That's
    >     not a very good solution. We changed this and put uucp and fax in the same
    >     group. That works...
    > 
    > The reason for uid(uucp) == uid(fax) are the lock files of UUCP.
    > The normal directory for UUCP-locking is /var/spool/locks,
    > owned by "uucp" with permission 0755. I just checked another
    > Linux box in my company (also a S.u.S.E. distribution)
    > but there seems to be a bug (don't know if this system still
    > matches the CD-ROM):
    
    The uucp-locking directory is /var/lock in our system...

Happy days with standards ... and:
$ uname -a
Linux wilson 2.0.14 #5 Thu Aug 22 17:01:26 MET DST 1996 i586
$ ls -ld /var/lock
drwxr-xr-x   3 root     root         1024 Mar 17  1995 /var/lock
$ 
     
    	....
    
    > If you set "fax" and "uucp" into the same group (e.g. "uucp")
    > you must (at least) also change the rights of this directory.
    > 
    > If mgetty+sendfax can live with any uid for "fax" why not
    > setting it to the same as "uucp" per default in S.u.S.E.?
    
    Some guys here weren't very happy with this solution, so I tried to give
    fax another uid, changed the faxaddmodem to give the FIFO.xxx file
    uucp.uucp as owner and all seems to work -all I have tested-. I use the
    faxserver at home, at our company and several other people use it, too.

I don't know your change and if this change may cause further
problems. There are a lot of complaints in the mailing-list like
"can't create/open FIFO.xxxx" .... This doesn't mean that
this is caused by your changes (a lot of them are not related to
Linux at all), I just want to say: changes like this may affect
other places too.

If you want to add mods to the upcoming pl2, please contact me
directly.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 10 Nov 1997 13:08:01 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Carsten Hoeger <choeger@suse.de>
Cc: flexfax@sgi.com
Subject: Re: flexfax: faxsetup on Linux (bug in moduser?)
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Mon, 10 Nov 1997, Carsten Hoeger wrote:

> On Mon, 10 Nov 1997, Nico Garcia wrote:
> 
> > On Mon, 10 Nov 1997, Carsten Hoeger wrote:
> > 
> > > Another point is, that the uid of uucp and fax has to be the same. That's
> > > not a very good solution. We changed this and put uucp and fax in the same
> > > group. That works...
> > 
> > Ummm. This is not ideal. Dial-up lines and lock files are normally
> > owned by the calling process, UUCP or HylaFAX. Using the same group
> > for the uucp/fax users should only work if the lock files are left
> > with group write permission, and the serial lines left owned by that
> > user. I'd want to think about that before doing it.
> 
> Why not rw-permissions for the 'dialin' groups lock-files?
> Do you think that this may be a security hole?

That's why I'd have to think about it. It requires the addition and
manipulation of a group permission, an otherwise unnecessary feature.
It also means that every other program in the universe, such as PPP,
minicom, and commercial packages would need to comply with a new
standard. They'd also have to be careful not to alter the permissions
when they exit their programs, leaving them at this new standard
rather than the classic "crw------- uucp uucp" permissions and
ownership.

It's feasible: is it *worth* it, and what else will it break?

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNGdNez/+ItycgIJRAQEu8AP/SJ6GSCWRxoLUdjsQx3Q/TkRfqbFxfYjo
zzxWptueFcebCNv+aqItxyox3MGe1ziUxqUp1MIHyxl4EimXqxbqQkfU7TrxSBlr
1+BSUxKeb5JyGlsZb67YrbJK/B5HjfyZ6pQVoWK9Xc5OfveDNwvE+nyuQwZoMHSy
M3VpllflyjM=
=Egqt
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 10 Nov 1997 14:36:22 -0500
From: Ramana Juvvadi <juvvadi@lekha.org>
To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
Cc: choeger@suse.de (Carsten Hoeger), flexfax@sgi.com
Subject: Re: flexfax: faxsetup on Linux (bug in moduser?)
Sender: owner-flexfax@celestial.com


On a related note, I ran into the following problem. On Redhat
Linux (probably on other Linuxes as well) uucp has a userid
of 10 and groupid of 14. The group wheel has the groupid of 10.

I saw that some files are created with a group of wheel, i.e.
group id of 10. I think there is a bug in the code somewhere
which mistakes user id with a group id. This may not have been
detected on systems which has same user and group ids.

Finally, if you are spending any time on Linux, I urge you to
do that Redhat or Slackware. Together these two distributions account 
for around 90% of the Linux users. 

Ramana 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
X-Authentication-Warning: tui.pinnacle.co.nz: jonc owned process doing -bs
Date: Tue, 11 Nov 1997 14:29:28 +1300 (NZDT)
From: Jonathan Chen <jonc@pinnacle.co.nz>
To: Ramana Juvvadi <juvvadi@lekha.org>
cc: Matthias Apitz <Matthias.Apitz@SOFTCON.de>,
        Hylafax Mailing list <flexfax@sgi.com>
Subject: Re: flexfax: faxsetup on Linux (bug in moduser?)
Sender: owner-flexfax@celestial.com

On Mon, 10 Nov 1997, Ramana Juvvadi wrote:

> 
> On a related note, I ran into the following problem. On Redhat
> Linux (probably on other Linuxes as well) uucp has a userid
> of 10 and groupid of 14. The group wheel has the groupid of 10.
> 
> I saw that some files are created with a group of wheel, i.e.
> group id of 10. I think there is a bug in the code somewhere
> which mistakes user id with a group id. This may not have been
> detected on systems which has same user and group ids.

This isn't actually a bug, HylaFAX has its own user-id (via
.../fax/etc/hosts) system which it maps onto UNIX's by using the
gid. ie HylaFAX uid => UNIX gid
--
Jonathan Chen <jonc@pinnacle.co.nz>                 Once is dumb luck.
                                                 Twice is coincidence.
             Three times and Somebody Is Trying To Tell You Something.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Tue, 11 Nov 1997 19:59:56 -0500
From: Ramana Juvvadi <juvvadi@lekha.org>
To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
Cc: raoul@cirl.meei.harvard.edu (Nico Garcia), flexfax@sgi.com
Subject: Re: flexfax: faxsetup on Linux (bug in moduser?)
Sender: owner-flexfax@celestial.com

Matthias.Apitz@SOFTCON.de writes:

 > The HylaFAX software can't work corretly if the user "fax"
 > in the passwd(4) has a different uid than "uucp". If there

I just want to clarify one thing on the user fax. I don't know why
there is already a user fax on S.U.S.E. There is no user
called fax on the standard Redhat distribution. And Redhat does come 
with mgetty+sendfax.

Ramana

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 10 Nov 1997 19:36:13 -0500 (EST)
From: Eddie Reed <eddie@eddiereed.com>
To: flexfax@sgi.com
Subject: flexfax: permissions
Sender: owner-flexfax@celestial.com

I get a password prompt when sending commands from the console (sendfax;
faxstat).  The only entries in /var/spool/fax/etc/hosts are "localhost"
and "127.0.0.1".
How can I set up the /etc/hosts file so anyone could send a fax from the
command prompt?
Regards
eddie

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
X-Authentication-Warning: tui.pinnacle.co.nz: jonc owned process doing -bs
Date: Tue, 11 Nov 1997 15:48:45 +1300 (NZDT)
From: Jonathan Chen <jonc@pinnacle.co.nz>
To: Eddie Reed <eddie@eddiereed.com>
cc: flexfax@sgi.com
Subject: Re: flexfax: permissions
Sender: owner-flexfax@celestial.com

On Mon, 10 Nov 1997, Eddie Reed wrote:

> I get a password prompt when sending commands from the console (sendfax;
> faxstat).  The only entries in /var/spool/fax/etc/hosts are "localhost"
> and "127.0.0.1".
> How can I set up the /etc/hosts file so anyone could send a fax from the
> command prompt?

Specify a specific user-id number for the entries in
/var/spool/fax/etc/hosts. Make sure there are no trailing white space
or comments.
--
Jonathan Chen <jonc@pinnacle.co.nz>                 Once is dumb luck.
                                                 Twice is coincidence.
             Three times and Somebody Is Trying To Tell You Something.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Problems sending Fax with USR Courier I-Modem
To: hylafax@libra.it (Guges-D)
Date: Mon, 10 Nov 1997 21:55:39 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Guges-D wrote:
    
    Hi, i just made my first mistake sending the subscribe request to the ML
    ^_^;;;;
    
    My problem is very easy, I have a USR Courier I-Modem v34/ISDN, it's an 2.0
    fax modem and I added it without no problem with the faxaddmodem program
    like a 2.0 fax modem. When I try to send a fax (sendfax) the modem do the
    call and when the recive fax start the comunications the USR hungup.
    This is a log of the call:
    
    Nov 10 10:38:29.71: [  902]: --> [22:AT+FCC=1,5,2,2,0,0,0,0]
    Nov 10 10:38:29.71: [  902]: --> [2:OK]
    Nov 10 10:38:29.71: [  902]: <-- [22:AT+FLI="NothingSetup"\r]
    Nov 10 10:38:29.85: [  902]: --> [21:AT+FLI="NothingSetup"]
    Nov 10 10:38:29.85: [  902]: --> [2:OK]
    Nov 10 10:38:29.85: [  902]: DIAL 0226268332
    Nov 10 10:38:29.85: [  902]: <-- [15:ATDT0226268332\r]
    Nov 10 10:38:53.49: [  902]: --> [14:ATDT0226268332]

Your modem/TA hybrid runs in AT-echo mode; this will not
work and you have to check why echo isn't turned off;

	matthias

    Nov 10 10:38:53.49: [  902]: SEND FAILED: Unknown problem (check modem power)
    Nov 10 10:38:53.49: [  902]: <-- [5:ATH0\r]
    Nov 10 10:38:53.49: [  902]: --> [4:+FCO]
    Nov 10 10:38:55.30: [  902]: --> [7:+FHS:02]
    Nov 10 10:38:55.30: [  902]: REMOTE HANGUP: Call aborted,  from +FK or
    <CAN> (code 2)
    Nov 10 10:38:55.30: [  902]: --> [2:OK]
    Nov 10 10:38:55.30: [  902]: SESSION END
    
    Anybody can help me ??
    
    If I use an Cardinal 28.8 fax modem (class 2) I have no problem. I tried to
    send the fax with the USR to faxmodem or nly fax machine bat I have the
    same problem
    
    thanks
    
    ---
    Stefano Rota                                      
    Resp.Tecnico                                      Tel./FAX 0341/493157
    Libra S.r.l.				       E-Mail hylafax@libra.it

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Tue, 11 Nov 1997 06:28:04 -0500 (EST)
From: Eddie Reed <eddie@eddiereed.com>
To: flexfax@sgi.com
Subject: flexfax: user ID discrepancy- fax vs. uucp
Sender: owner-flexfax@celestial.com

Using BSDI 3.0, I get an error message "duplicate user id fax 6 uucp 6"
 There has been a considerable amount of correspondence on this
list regarding a similar problem with Linux.  What is the final solution
on solving this problem? Assign the user fax a different ID in
/etc/master.passwd, but preserve gid "6"?
REgards
Eddie

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: user ID discrepancy- fax vs. uucp
To: eddie@eddiereed.com (Eddie Reed)
Date: Tue, 11 Nov 1997 21:04:13 +0000 (GMT)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> 
> Using BSDI 3.0, I get an error message "duplicate user id fax 6 uucp 6"

If this is more than a warning, I'd get BSDI fixed.  Having multiple
aliases for root, in particular, is safer than making several people
share the same password.

Incidentally, there are other areas where it is unwise to blindly accept
the Linux user id structure, e.g. unless you explicitly set the default
user for sendmail, it will not use the, correct, daemon, on Slackware linux.
The user it does use is not one that you should want it to use for 
marginally trusted operations.  I get the impression, also that some of
the user IDs in Slackware were put there without any real understanding.

My suspicion is that the correct fix, if it really becomes necessary
is to globally replace fax by uucp, whenever a user-ID is used within
Hylafax.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Tue, 11 Nov 1997 22:25:14 -0800 (PST)
From: Tim Rice <tim@trr.metro.NET>
Reply-To: Tim Rice <tim@trr.metro.NET>
To: Matthias Apitz <matthias.apitz@softcon.de>
cc: HylaFax mailing list <flexfax@sgi.com>
Subject: flexfax: package patch
Sender: owner-flexfax@celestial.com


Here is a patch I hope to see included in pl2

It allows SVR4 machines to do a "make package" to produce packages
that can be installed with the pkgadd(1M) command. It makes a
client only package and a full package.

It customizes the package based on the results of configure.

Since HylaFAX requires the tiff tools, I have also included a patch
for tiff-v3.4beta035 to make it into a package.

I have only tested this on UnixWare 2.03 and 2.1.2

Included are 4 files.	hylapkg.patch hylapkg.tar.gz
			tiffpkg.patch tiffpkg.tar.gz

To apply the patch,
$ cd /your-path/tiff-v3.4beta035
$ patch < tiffpkg.patch
$ gzcat tiffpkg.tar.gz | tar xovf -
$ ./configure
$ make package

Now for HylaFAX
$ cd /your-path/hylafax-v4.0pl1
$ patch < hylapkg.patch
$ gzcat hylapkg.tar.gz | tar xovf -
$ ./configure

Now read the README in the pkg directory.
$ make package


Enjoy.

--
Tim Rice			Multitalents	(707) 887-1469 (voice)
tim@trr.metro.net		Keeper of the HylaFAX SCO binaries
*** Makefile.in.orig	Tue Nov 26 15:20:19 1996
--- Makefile.in	Tue Nov 11 21:23:16 1997
***************
*** 183,188 ****
--- 183,189 ----
  	else \
  	    true; \
  	fi
+ 	@${ECHO} "= "pkg; cd pkg; ${MAKE} clean
  clobber::
  	(cd util; ${MAKE} clobberconfig)
  	(cd etc; ${MAKE} clobberconfig)
***************
*** 236,241 ****
--- 237,243 ----
  	else \
  	    true; \
  	fi
+ 	(cd pkg; ${MAKE} distclean)
  	-${RM} -f defs rules Makefile
  
  makeClientDirs::
***************
*** 361,366 ****
--- 363,371 ----
  	else \
  	    true; \
  	fi
+ 
+ package::
+ 	@${ECHO} "= "pkg; cd pkg; ${MAKE} package
  
  product::
  	test -d dist || ${MKDIR} dist
*** config.site.orig	Tue Nov 26 15:20:18 1996
--- config.site	Tue Nov 11 21:23:16 1997
***************
*** 102,107 ****
--- 102,113 ----
  #PATH_IMPRIP="/usr/lib/print/psrip"	# pathname of Impressario 2.1 RIP
  #MANSCHEME="sysv-source-cat-strip"	# manual page installation scheme
  
+ # SVR4 packaging stuff
+ #PKG_ARCH=				# ARCH variable in pkginfo file
+ #PKG_EMAIL=someone@somehost.somedomain	# EMAIL variable in pkginfo file
+ #PKG_VENDOR="Your Name Here"		# VENDOR variable in pkginfo file
+ #
+ 
  #
  # Parameters used when building the software.
  #
*** configure.orig	Tue Nov 26 15:20:17 1996
--- configure	Tue Nov 11 21:23:16 1997
***************
*** 81,86 ****
--- 81,91 ----
  LIBDB="-L/usr/local/lib -ldb"
  DBLIBINC=/usr/local/include
  
+ # SVR4 packaging stuff
+ PKG_ARCH=				# ARCH variable in pkginfo file
+ PKG_EMAIL=someone@somehost.somedomain	# EMAIL variable in pkginfo file
+ PKG_VENDOR="Your Name Here"		# VENDOR variable in pkginfo file
+ 
  : ${MAKE=make}				# make to use
  # screws up the test of `-f -'
  unset MAKEFLAGS
***************
*** 185,190 ****
--- 190,198 ----
  PATHGETTY	PATH_GETTY
  PATHVGETTY	PATH_VGETTY
  PATHEGETTY	PATH_EGETTY
+ PKG_ARCH
+ PKG_EMAIL
+ PKG_VENDOR
  PORT
  PORTFUNCS
  PSPACKAGE	PS
***************
*** 3280,3285 ****
--- 3288,3298 ----
  CPU=`echo $TARGET | sed 's/-.*//'`
  Note "Using $FILLORDER bit order for your $CPU cpu."
  
+ # SVR4 packaging stuff
+ if [ -z "$PKG_ARCH" ]; then
+ 	PKG_ARCH=$CPU
+ fi
+ 
  #
  # Style of getty support.
  #
***************
*** 4126,4131 ****
--- 4139,4153 ----
  # util/recvstats.sh	script for receive statistics accounting
  # util/faxcron.sh	script for cleaning up the spooling area (run from cron)
  # util/pagesizes	client page dimensions database (has default page size)
+ # pkg/cpkginfo		SVR4 package stuff (client package info)
+ # pkg/cproto.stub	SVR4 package stuff (client prototype stub)
+ # pkg/crequest		SVR4 package stuff
+ # pkg/make_proto.sh	SVR4 package stuff (script for making prototype files)
+ # pkg/postinstall	SVR4 package stuff
+ # pkg/postremove	SVR4 package stuff
+ # pkg/spkginfo		SVR4 package stuff (server package info)
+ # pkg/sproto.stub	SVR4 package stuff (server prototype stub)
+ # pkg/srequest		SVR4 package stuff
  #
  CONF_FILES="
      defs
***************
*** 4159,4164 ****
--- 4181,4197 ----
      util/faxcron.sh
  
      util/pagesizes
+ 
+     pkg/Makefile
+     pkg/cpkginfo
+     pkg/cproto.stub
+     pkg/crequest
+     pkg/make_proto.sh
+     pkg/postinstall
+     pkg/postremove
+     pkg/spkginfo
+     pkg/sproto.stub
+     pkg/srequest
  "
  SedConfigFiles $CONF_FILES
  test $DPS = yes &&		   SedConfigFiles dps/Makefile
Attachment Converted: "C:\PROGRAM FILES\EUDORA\Attach\hylapkg.tar.gz"
*** Makefile.in.old	Mon Jun 10 15:24:57 1996
--- Makefile.in	Thu Sep 25 09:04:18 1997
***************
*** 72,77 ****
--- 72,78 ----
  	@${ECHO} "= "libtiff; cd libtiff; ${MAKE} clean
  	@${ECHO} "= "tools; cd tools; ${MAKE} clean
  	@${ECHO} "= "man; cd man; ${MAKE} clean
+ 	@${ECHO} "= "pkg; cd pkg; ${MAKE} clean
  #	-cd contrib/dbs; ${MAKE} clean
  #	-cd contrib/dbs/xtiff; ${MAKE} clean
  
***************
*** 82,87 ****
--- 83,93 ----
  	rm -f man/Makefile
  	rm -f port/Makefile port/install.sh
  	rm -f html/Makefile
+ 	rm -f pkg/Makefile pkg/make_proto.sh pkg/pkginfo
+ 	rm -f pkg/prototype.stub pkg/request
+ 
+ package:
+ 	@${ECHO} "= "pkg; cd pkg; ${MAKE} package
  
  #
  # The folllowing rule creates a binary distribution for IRIX.
*** config.site.old	Wed Oct 30 12:23:22 1996
--- config.site	Tue Sep 23 21:59:44 1997
***************
*** 113,122 ****
--- 113,130 ----
  #DIR_GZLIB="../zlib"			# directory for libz
  
  #
+ # SVR4 packaging stuff
+ #SYSUID=bin				# user ID for pkgadd installs
+ #SYSGID=				# group ID for pkgadd installs
+ #PKG_ARCH=				# ARCH variable in pkginfo file
+ #PKG_EMAIL=someone@somehost.somedomain	# EMAIL variable in pkginfo file
+ #PKG_VENDOR="Your Name Here"		# VENDOR variable in pkginfo file
+ 
  # Miscellaneous parameters.
  #
  #FILLORDER="MS2LSB"			# bit order of cpu (MSB2LSB/LSB2MSB)
  #MANSCHEME="sysv-source-cat-strip"	# manual page installation scheme
+ #
  
  #
  # Parameters used when building the software.
*** configure.old	Mon Jun 10 15:24:56 1996
--- configure	Wed Sep 24 20:55:23 1997
***************
*** 69,74 ****
--- 69,81 ----
  : ${MAKEFLAGS=}				# unset MAKEFLAGS
  RM="rm -f"
  
+ # SVR4 packaging stuff
+ SYSUID=bin				# user ID for pkgadd installs
+ SYSGID=					# group ID for pkgadd installs
+ PKG_ARCH=				# ARCH variable in pkginfo file
+ PKG_EMAIL=someone@somehost.somedomain	# EMAIL variable in pkginfo file
+ PKG_VENDOR="Your Name Here"		# VENDOR variable in pkginfo file
+ 
  #
  # Error diagnostics that should go to the terminal are
  # done with this interface (or cat).
***************
*** 221,226 ****
--- 228,238 ----
  fi
  exec 5>./config.log			# compiler messages and the like
  
+ Note()
+ {
+     echo "$@" 1>&4
+ }
+ 
  capture()
  {
      (eval "set -x; $*") >&5 2>&1
***************
*** 1268,1273 ****
--- 1280,1315 ----
  # and may be interactively changed.
  #
  
+ #
+ # System installation group ID.
+ #
+ if [ -z "$SYSGID" ]; then
+     case $TARGET in
+     *-aix*)	SYSGID=sys;;
+     *bsd*)	SYSGID=bin;;
+     *-hpux*)	SYSGID=bin;;
+     *-irix*)	SYSGID=sys;;
+     *-isc*)	SYSGID=sys;;
+     *-linux*)	SYSGID=bin;;
+     *-sco*)	SYSGID=sys;;
+     *-solaris*)	SYSGID=sys;;
+     *-sunos*)	SYSGID=bin;;
+     *-sysv4*)	SYSGID=sys;;
+     *-ultrix*)	SYSGID=bin;;
+     *)	
+ 	SYSGID=sys
+ 	for i in sys bin; do
+ 	    grep "^${i}:" /etc/group >/dev/null 2>&1 && { SYSGID=$i; break; }
+ 	done
+ 	;;
+     esac
+ fi
+ Note "Using uid $SYSUID and gid $SYSGID for packaging programs."
+ 
+ if [ -z "$PKG_ARCH" ]; then
+ 	PKG_ARCH=`echo $TARGET | sed 's/-.*//'`
+ fi
+ 
  Note ""
  Note "Selecting default TIFF configuration parameters."
  Note ""
***************
*** 1734,1739 ****
--- 1776,1784 ----
  /@MANSEDLOCAL@/s;;${MANSEDLOCAL};g
  /@MANLIB@/s;;${MANLIB};g
  /@MV_F@/s;;${MV_F};g
+ /@PKG_ARCH@/s;;${PKG_ARCH};g
+ /@PKG_EMAIL@/s;;${PKG_EMAIL};g
+ /@PKG_VENDOR@/s;;${PKG_VENDOR};g
  /@PORT@/s;;${PORT};g
  /@PORTFUNCS@/s;;${PORTFUNCS};g
  /@RANLIB@/s;;${RANLIB};g
***************
*** 1743,1748 ****
--- 1788,1795 ----
  /@SRCDIR@/s;;${SRCDIR};g
  /@RELSRCDIR@/s;;`relativize ${SRCDIR}`;g
  /@STRIP@/s;;${STRIP};g
+ /@SYSGID@/s;;${SYSGID};g
+ /@SYSUID@/s;;${SYSUID};g
  /@TARGET@/s;;${TARGET};g
  /@TIFFLIBREF@/s;;${TIFFLIBREF};g
  /@VERSION@/s;;${VERSION};g
***************
*** 1814,1819 ****
--- 1861,1873 ----
      test $HTML = yes &&		CONF_FILES="$CONF_FILES html/Makefile"
      test $PORT = yes &&		CONF_FILES="$CONF_FILES port/Makefile"
  fi
+ CONF_FILES="$CONF_FILES
+ 	pkg/Makefile
+ 	pkg/make_proto.sh
+ 	pkg/pkginfo
+ 	pkg/prototype.stub
+ 	pkg/request
+ 	"
  SedConfigFiles $CONF_FILES
  
  Note "Done."
Attachment Converted: "C:\PROGRAM FILES\EUDORA\Attach\tiffpkg.tar.gz"

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Re: HylaFAX
To: juvvadi@lekha.org (Ramana Juvvadi)
Date: Thu, 13 Nov 1997 15:05:32 +0100 (MEZ)
Cc: flexfax@sgi.com, raoul@cirl.meei.harvard.edu (Nico Garcia)
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Ramana Juvvadi wrote:
    
    Nico Garcia writes:
    
     > in the default configuration and local configuration files. You might
     > check the ownership and permissions of /var/spool/fax: I seem to
     > remember that you are using a uid for the "fax" user that is not the
     > same as the "uucp" user, and having fun tracking down all the things
     > that this breaks?
    
    
    If the user fax already exists the rpm installation doesn't modify
    it. The assumption is that the system administrator knows what
    he is doing with the user fax. 

The assumption of the HylaFAX software is that the uid
of "uucp" and "fax" are the same and that the script
faxsetup(1M) detects any conflict for an existing user
"fax" or creates one if "fax" doesn't exist.

Letting the thing go on with an invalid uid for "fax" will lead
to problems in HylaFAX which most of the Linux users don't understand
and could not resolve. The bottom line of this fault (not running
or wrong changing faxsetup(1M)) is the mail traffic in this list
aka "I'm new to it ... and it doesn't work"

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 13 Nov 1997 09:14:17 -0500 (EST)
From: Helen Birkmann <hbirkmann@mkl.com>
To: Jonathan Chen <jonc@pinnacle.co.nz>
cc: Eddie Reed <eddie@eddiereed.com>, flexfax@sgi.com
Subject: Re: flexfax: permissions
Sender: owner-flexfax@celestial.com

On Tue, 11 Nov 1997, Jonathan Chen wrote:

> On Mon, 10 Nov 1997, Eddie Reed wrote:
> 
> > I get a password prompt when sending commands from the console (sendfax;
> > faxstat).  The only entries in /var/spool/fax/etc/hosts are "localhost"
> > and "127.0.0.1".
> > How can I set up the /etc/hosts file so anyone could send a fax from the
> > command prompt?
> 
> Specify a specific user-id number for the entries in
> /var/spool/fax/etc/hosts. Make sure there are no trailing white space
> or comments.

In other words, just list users such as 
www
84
nobody
32766
nogroup
32767

??

Regards
Eddie

> --
> Jonathan Chen <jonc@pinnacle.co.nz>                 Once is dumb luck.
>                                                  Twice is coincidence.
>              Three times and Somebody Is Trying To Tell You Something.
> 
> 

************************************************************************
*	MKL.Net		        	Internet Service Provider      *
*	A service of CPU Industries                                    *
*	Bainbridge, New York, USA      	PPP, dial-up, telnet, uucp,    *
*	http://www.mkl.com		virtual domains, web design    *
*	https://mall.mkl.com	Secured server - Merchant services     *
************************************************************************

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Sun, 09 Nov 1997 19:00:11 +0000
From: "Thorbjrn Eriksson" <pt96ter@student.hk-r.se>
To: JK Lavacek <jlav@cpicorp.com>, flexfax@sgi.com
Subject: Re: flexfax: PCL to postscript?
X-MIME-Autoconverted: from 8bit to quoted-printable by asterix.rby.hk-r.se id SAA04272
Sender: owner-flexfax@celestial.com

JK Lavacek wrote:

> Has anyone had experience/success with a conversion of pcl formatted
> files into postscript?
> Seems the format of files recieved onto the faxserver, will be in pcl
> format.  Due to things
> beyond my control, I have to live with this.  Attempted to compile a
> source distribution of
> a file called "lj2ps".  This gives various errors in the makefile.  With
> a limited knowledge
> of C, I'm lost.  Seems the line endind \ is confusing my version of
> gcc...? Removed a few \
> and joined lines in some .c files and the error moved down in lines.
> Any help would be greatly
> appreciated.  Thanks in advance,
>                         Jason Lavacek <jlav@cpicorp.com>
>
> HYLAfax version:  4.0 patch level 1
> Linux (redhat 4.2) kernel 2.0.30

I've been using lj2ps for a while together with HylaFAX and, after some
minor changes, lj2psworks fine for me. It was a couple of months since I
made it work, so I can't remember what problems
I had compiling it.

It seems like we use the same HylaFAX distribution and Linux kernel, but I
dont know if the
gcc version is the same.

If you would like me to, I could send you the sourcefiles and makefile, but
the changes I've made
are not supposed to be any official patch or something like that. Ive just
made it work for me.

Thorbjrn Eriksson


>

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Tue, 11 Nov 1997 12:15:23 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Chemisov Sergey <sysadm@ns.sirena.ru>
Cc: flexfax@sgi.com
Subject: Re: flexfax: faxserver to faxserver
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Tue, 11 Nov 1997, Chemisov Sergey wrote:

>                         Good  day !
>         Explain for me please :
> Can I configure my HylaFAX-server that faxes which comes to him 
> from local fax-machine(from paper-document), sends to anover
> HylaFAX-server,and after goes 
> to remote fax-machine(to paper-document) ?
> 
> Chemisov Serg.

Yes, sort of. When the fax is received, a script called "faxrcvd" gets
run. This script can do whatever you tell it to do within limits.
It's very possible to set it to forward all faxes to another fax
machine, such as the second HylaFAX server you mention. That second
machine would then have to modify its "faxrcvd" to print all faxes.

Alternatively, if the machines exist on the same local network, it's
possible to transmit the faxes as print jobs for local printing on the
second server, and not use HylaFAX there at all.  Hopefully this
covers your situation....

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNGiSBD/+ItycgIJRAQGyswQAgofgvlkfVWXYIBY9QuUge9RBaGyFsLMK
p32HLd8RCLwmbjDolETzO19GtrLHkBhwN+ZMxHIU9LLDSigGC3EELGqpePz4ws4D
4En3MrbHA94rEGKQP09UORsHNA4yUxo+5akK3qKPi6QrFaBJt6+UxJSYkBIiMVx+
8jbnNjJ23AY=
=ySaK
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Comments: Authenticated sender is <spine001@gold.tc.umn.edu>
From: Julio C Spinelli <spine001@gold.tc.umn.edu>
To: Julio C Spinelli <spine001@gold.tc.umn.edu>, flexfax@sgi.com
Date: Tue, 11 Nov 1997 23:23:00 +0000
Subject: flexfax: To do list
Sender: owner-flexfax@celestial.com

REmember to change the ability of the fax to receive files in the 
remote systems. When the receiving fax machine has that ability, the 
risk is that a customer may send a file in this format and then we 
won't be able to convert it back to a standard tiffg format in a 
lower resolution

(vr),(br),(wd),(ln),(df),(ec),(bf),(st)

bf must be zero to disable this,

Julio

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Wed, 12 Nov 1997 11:29:16 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Michael Hartges <Michael.Hartges@Henkelhausen.com>
Cc: flexfax@sgi.com
Subject: Re: flexfax: EPSON -> FAX
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Wed, 12 Nov 1997, Michael Hartges wrote:

> Hi,
> 
> i4m looking for a prg, which converts textfiles with
> EPSON-controlsequences to a FAX-format automatically on my SCO OS5-host.
> 
> tia Hardy

Ghostscript is your friend. Take a look at the fax2ps scripts as part
o HylaFAX for ideas on how to write such commands.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNGnZVD/+ItycgIJRAQFf/wP+KdQNRkqQzxTBN4tIssXqtxC5bn+gIbMN
CkEOzToB+BOpCowVFyX8uaO/mx04P+fjrrud3PAN0XG58b91vO8X2YxeZM3bqyan
Ot6gLZeSaLcwxBry0v1M7OP2BBWyA6HFpHmt3bKS+z7oi9by+T0krepyFB2YZwAe
OpddVcq0Ol8=
=Vo+k
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Alexandre Meissonnier <ameissonnier@dohle.com>
Reply-To: "ameissonnier@dohle.com" <ameissonnier@dohle.com>
To: "HylaFAX-Mailingliste (E-Mail)" <flexfax@sgi.com>
Subject: flexfax: fax2ps again
Date: Wed, 12 Nov 1997 17:26:16 +0100
Organization: Dohle Handelsgruppe Systemberatung GmbH
Sender: owner-flexfax@celestial.com

I want to run HylaFAX on my Linux box running S.u.S.E.-Linux-5.0.0#0. The 
problem is that the fax2ps program always crashes with a "Segmentation 
fault". Now I got the source of this program, compiled it and - the result 
is the same. No way to convert a received fax into PostScript.

What can I do to make the program work? Has anybody got the same problem?


	Alexandre

------------------------------------------------------------------------  
-------
Alexandre Meissonnier
Dohle Handelsgruppe Systemberatung GmbH
53719 Siegburg (Germany)
ameissonnier@dohle.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Wed, 12 Nov 1997 18:19:49 -0800
From: "Luiz Otavio L. Zorzella" <zorzella@iname.com>
To: flexfax@sgi.com
Subject: flexfax: aborting faxes
Sender: owner-flexfax@celestial.com


Hi,

I'm trying to share my modem between hylafax and my pppd.

I'm being able to do it using the "faxstate" command: I issue a
"faxstate -s busy" before I call pppd, so that hylafax won't mess with
my connection, and a "faxstate -s ready" after the pppd goes down, so
that the waiting faxes will be sent. 

My problem is that I'm trying to make this thing automatic (putting
this commands in my scripts), but I'm having problems when the pppd
trys to "go up", and hylafax is using the modem.

Even though my script does "faxstate -s ready", and sends
initialization strings to my modem, hylafax insists on keeping using
the modem, and then neither pppd nor hylafax are able to do
anything... Seems like hylafax ignores the status of the modem if it
is already trying to send something (i.e., while there are still faxes
waiting to be sent in sendq).

So I thought maybe I should have to issue some abort command to
hylafax, but I could not figure out anything.

Anyone has any idea?

-- 
Z  .:.   .:.   Luiz Otavio Lautenschlaeger Zorzella       .:.   .:.
  :   : :   :                                            :   : :   :
  : .:: : .::  mailto:zorzella@iname.com                 ::. : ::. :
  : ::: : :::  http://conexware.cps.softex.br/~zorzella  ::: : ::: :
  `.::' `.::'  PGP at my home page                       `::.' `::.'

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 13 Nov 1997 18:21:39 +0100 (MET)
From: Carsten Hoeger <choeger@suse.de>
To: Olaf Menzel <menzel@tfh-berlin.de>
cc: flexfax@sgi.com
Subject: Re: flexfax: SuSE 5.0 & HylaFAX
Sender: owner-flexfax@celestial.com

On Thu, 13 Nov 1997, Olaf Menzel wrote:

> Hello,
> ------
> I am really a HylaFAX newcomer and have some questions.
> 
> I use HylaFax from the SuSE 5.0 distribution and I have completed
> successfully faxsetup & faxaddmodem. Then I have set the right in the file
> hosts in /var/spool/fax/etc.
> 
> When I check the server with faxstat -srv everything seemes to be o.k. but
> if I try to fax a file with sendfax the fax ist only queued and nothing else
> happens.
> 
> If I send the file test.ps for example to the number 123456789
> with the command :
> 
> sendfax -n -d 123456789 test.ps
> 
> I have got the following answer from faxstat -s:
> 
>    HylaFAX scheduler on faxserv.somewhere.else: Running
> 
>    JID  Pri S  Owner Number       Pages Dials     TTS Status
>    2    127 W menzel 123456789    0:0   0:12         

[...]

You have either to start faxgetty -just comment out the entry in
/etc/inittab- or tell the queuer with faxmodem about its fax-related
capabilities.

See http://www.suse.de/Support/sdb/choeger_fax-sendonly.html
and http://www.suse.de/Support/sdb/choeger_hylafax-problems.html

I think in most situations, these URLs should help...

mfG,
	Carsten Hoeger
------
Carsten Hoeger  - S.u.S.E. GmbH -  Gebhardtstr. 2  -  90762 Fuerth  -  Germany
fax +49-911-3206727      web http://www.suse.de        mailbox +49-911-3247122
"Ein Buch ist eine Wolke, auf der man keinen Husten bekommt." (Max Goldt)

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: aborting faxes
To: zorzella@iname.com (Luiz Otavio L. Zorzella)
Date: Thu, 13 Nov 1997 16:10:18 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Luiz Otavio L. Zorzella wrote:
    
    
    Hi,
    
    I'm trying to share my modem between hylafax and my pppd.

I'm doing exact the same (HylaFAX && pppd) at home and in
my company....
    
    I'm being able to do it using the "faxstate" command: I issue a
    "faxstate -s busy" before I call pppd, so that hylafax won't mess with
    my connection, and a "faxstate -s ready" after the pppd goes down, so
    that the waiting faxes will be sent. 

... and they are working fine with the lock files in the
locking directory (/var/spool/lock, style ASCII); the
daemon faxgetty(1M) goes silently away if the modem is
used for outbound PPP traffic and send operations don't disturb
the PPP connections.

	matthias
    
    My problem is that I'm trying to make this thing automatic (putting
    this commands in my scripts), but I'm having problems when the pppd
    trys to "go up", and hylafax is using the modem.
    
    Even though my script does "faxstate -s ready", and sends
    initialization strings to my modem, hylafax insists on keeping using
    the modem, and then neither pppd nor hylafax are able to do
    anything... Seems like hylafax ignores the status of the modem if it
    is already trying to send something (i.e., while there are still faxes
    waiting to be sent in sendq).
    
    So I thought maybe I should have to issue some abort command to
    hylafax, but I could not figure out anything.
    
    Anyone has any idea?
    
    -- 
    Z  .:.   .:.   Luiz Otavio Lautenschlaeger Zorzella       .:.   .:.
      :   : :   :                                            :   : :   :
      : .:: : .::  mailto:zorzella@iname.com                 ::. : ::. :
      : ::: : :::  http://conexware.cps.softex.br/~zorzella  ::: : ::: :
      `.::' `.::'  PGP at my home page                       `::.' `::.'

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 13 Nov 1997 11:22:07 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: "Luiz Otavio L. Zorzella" <zorzella@iname.com>
Cc: flexfax@sgi.com
Subject: Re: flexfax: aborting faxes
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Wed, 12 Nov 1997, Luiz Otavio L. Zorzella wrote:

> Hi,
> 
> I'm trying to share my modem between hylafax and my pppd.
> 
> I'm being able to do it using the "faxstate" command: I issue a
> "faxstate -s busy" before I call pppd, so that hylafax won't mess with
> my connection, and a "faxstate -s ready" after the pppd goes down, so
> that the waiting faxes will be sent. 
> 
> My problem is that I'm trying to make this thing automatic (putting
> this commands in my scripts), but I'm having problems when the pppd
> trys to "go up", and hylafax is using the modem.

This should be unnecessary. HylaFAX and uucp and kermit and tip and
minicom and pppd should all be looking in the same directory for the
same style of lock files: this is /var/lock on Linux, /var/spool/locks
on SunOS, etc.

Try running each program and seeing where and what kind of lock file it
generates: these really should interoperate.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNGspKD/+ItycgIJRAQG5fAP/asAqzW+jTPszKybZMXGIVQSWxkBQEpPa
Zo7fPBAmmYPDE2gZbZ9x8Apm5Q/H3KsLsrYsZl66Hi0S9e6kNVDgw7yfxIl5kih3
36T4RB3uO/juZ1E7x9sHFWeVXIkW2qUtsT20uvmM27klbIo+e59Fie3B8AccCKll
k6fo29xIv74=
=9Phv
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 13 Nov 1997 09:35:28 -0500 (EST)
From: Eddie Reed <eddie@eddiereed.com>
To: flexfax@sgi.com
Subject: flexfax: Hylafax - fax user defined in passwd
Sender: owner-flexfax@celestial.com

In the process of getting Hylafax running, I get the following error in
the daily run output:

"Checking the /etc/master.passwd file:
 Login fax is off but still has a valid shell."


"/etc/master.passwd has duplicate user id's.
fax 6	uucp 6"

I changed the following user ID to #8 in the /etc/master.passwd file.

"Checking home directories:
user fax home directory is owned by uucp"

I changed the ownership of the home directory from user uucp to user fax

Now I can finally issue the command sendfx without getting prompted by a
password, but nothing happens.  The command faxstat is not returning any
output.

REgards
Eddie

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 13 Nov 1997 11:35:21 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Eddie Reed <eddie@eddiereed.com>
Cc: flexfax@sgi.com
Subject: Re: flexfax: Hylafax - fax user defined in passwd
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Thu, 13 Nov 1997, Eddie Reed wrote:

> In the process of getting Hylafax running, I get the following error in
> the daily run output:
> 
> "Checking the /etc/master.passwd file:
>  Login fax is off but still has a valid shell."
> 
> 
> "/etc/master.passwd has duplicate user id's.
> fax 6	uucp 6"
> 
> I changed the following user ID to #8 in the /etc/master.passwd file.

Bad Linux Admin! No biscuit! For various reasons, uucp and fax really
need to share the same user ID. This has been the subject of much
discussion lately, but it boils down to "making them different will
screw up handing the modem from one program to the other, and will
be much, much too hard to do otherwise."

> I changed the ownership of the home directory from user uucp to user fax
> 
> Now I can finally issue the command sendfx without getting prompted by a
> password, but nothing happens.  The command faxstat is not returning any
> output.

Simply put, you killed it. (No blame, just letting you know.) Change
this back to the same UID as uucp.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNGssQz/+ItycgIJRAQEaOwP/UXd4T/pheTvp0L3D1lHCzDaU2N1Pba6U
NM/GWT72wG+u7eC0AY26G2FpOEjK9BS4n+EIPOtlmqtvPj6BqdW6O576CCnMw5oC
sT1OvRFkGoPGBTJVugUPOabl8p7Y0WerDEL0BJs7BVzJHRNtHAgIJADi5ALNKgDA
bzt4AdIOJ+Y=
=eiUS
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 13 Nov 1997 14:01:26 -0500 (EST)
From: Eddie Reed <eddie@eddiereed.com>
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
cc: flexfax@sgi.com
Subject: Re: flexfax: Hylafax - fax user defined in passwd
Sender: owner-flexfax@celestial.com



On Thu, 13 Nov 1997, Nico Garcia wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> 
> On Thu, 13 Nov 1997, Eddie Reed wrote:
> 
> > In the process of getting Hylafax running, I get the following error in
> > the daily run output:
> > 
> > "Checking the /etc/master.passwd file:
> >  Login fax is off but still has a valid shell."
> > 
> > 
> > "/etc/master.passwd has duplicate user id's.
> > fax 6	uucp 6"
> > 
> > I changed the following user ID to #8 in the /etc/master.passwd file.
> 
> Bad Linux Admin! No biscuit! For various reasons, uucp and fax really
> need to share the same user ID. This has been the subject of much
> discussion lately, but it boils down to "making them different will
> screw up handing the modem from one program to the other, and will
> be much, much too hard to do otherwise."
> 
> > I changed the ownership of the home directory from user uucp to user fax
> > 
> > Now I can finally issue the command sendfx without getting prompted by a
> > password, but nothing happens.  The command faxstat is not returning any
> > output.
> 
> Simply put, you killed it. (No blame, just letting you know.) Change
> this back to the same UID as uucp.
> 
> 			Nico Garcia
> 			Engineer, CIRL 
> 			Mass. Eye and Ear Infirmary
> 			raoul@cirl.meei.harvard.edu
> 
> -----BEGIN PGP SIGNATURE-----
> Version: 2.6.2
> 
> iQCVAwUBNGssQz/+ItycgIJRAQEaOwP/UXd4T/pheTvp0L3D1lHCzDaU2N1Pba6U
> NM/GWT72wG+u7eC0AY26G2FpOEjK9BS4n+EIPOtlmqtvPj6BqdW6O576CCnMw5oC
> sT1OvRFkGoPGBTJVugUPOabl8p7Y0WerDEL0BJs7BVzJHRNtHAgIJADi5ALNKgDA
> bzt4AdIOJ+Y=
> =eiUS
> -----END PGP SIGNATURE-----
> 
> 
Yup, I darn killed it. And then erased all configuration files created
during install, then reinstalled hylafax from the distribution.
Ran faxsetup, ran faxmodem, back to abnormal:

Can't use sendfax from command prompt without a password.

The strange thing is that I had this working with the very first install!!
What threw me off is that snnp takes the same port as stronghold.  So I
reinstalled hylafax without the snnp and moved stronghold to another port.

When I look at the dir with the faxes that went out previously, they
report as being sent as "nobody@myhostname.com".  Which tells me there is
some kind of permissions problem here with some of the dirs and files
owned by "fax".

REgards
Eddie

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: Hylafax - fax user defined in passwd
To: eddie@eddiereed.com (Eddie Reed)
Date: Thu, 13 Nov 1997 20:16:47 +0000 (GMT)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> 
> "/etc/master.passwd has duplicate user id's.
> fax 6	uucp 6"

What operating system are you using.  From other reports, rejecting
aliased user ids seems to be a "feature" recently introduced into 
BSDI 386.

My advice would be to turn off the daily check program, and reset the
fax user id to be the same as uucp, but you should also read the archives
of the list (I assume they exist) as variations on this theme have been
common for the last couple of weeks.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 13 Nov 1997 16:36:27 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Eddie Reed <eddie@eddiereed.com>
Cc: flexfax@sgi.com
Subject: Re: flexfax: Hylafax - fax user defined in passwd
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Thu, 13 Nov 1997, Eddie Reed wrote:

> Yup, I darn killed it. And then erased all configuration files created
> during install, then reinstalled hylafax from the distribution.
> Ran faxsetup, ran faxmodem, back to abnormal:
> 
> Can't use sendfax from command prompt without a password.

OK. This is a separate problem, methinks. Did you re-run faxsetup?
Does "faxstat" work? Is your $SPOOL/etc/hosts file for HylaFAX
permissions set up correctly, which may be the problem?

> The strange thing is that I had this working with the very first install!!
> What threw me off is that snnp takes the same port as stronghold.  So I
> reinstalled hylafax without the snnp and moved stronghold to another port.

Oof!. That's a potetntial problem for the rest of us: thanks for the
heads up. Is 444 the default port for stronghold?

> When I look at the dir with the faxes that went out previously, they
> report as being sent as "nobody@myhostname.com".  Which tells me there is
> some kind of permissions problem here with some of the dirs and files
> owned by "fax".

Umm. No. hfaxd sits on the port 4559, and answers requests for service
from the sendfax or faxstat command. They correspond, the user's name
gets associated by hfaxd with the service request (using permissions
in the hosts file to verify their authority to send faxes) and an
email address gets generated from that data (as I understand it). The
data gets fed to the hfaxd by sendfax, and it handles the fax job.  No
ownership or file permissions involved there.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNGty1D/+ItycgIJRAQEaWAP/YQ+0m/L9oaiLx6IB/6Az2tTcm3rezDrH
zz4PTg6OVAle6VrfkwFKRA1gqsLPuUtfk7wB5q8bj14pqRXpoBtvB/CB9q3zAmd7
WVdxs8sc818l1TOSjA6Huw8goZCFRp6Yxne8mRRr1X6Z+moI8cJkcYdRYalHF82m
x+056fJ5FsA=
=/uvk
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 13 Nov 1997 18:25:26 -0500 (EST)
From: Eddie Reed <eddie@eddiereed.com>
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
cc: flexfax@sgi.com
Subject: Re: flexfax: Hylafax - fax user defined in passwd
Sender: owner-flexfax@celestial.com


On Thu, 13 Nov 1997, Nico Garcia wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> 
> On Thu, 13 Nov 1997, Eddie Reed wrote:
> 
> > Yup, I darn killed it. And then erased all configuration files created
> > during install, then reinstalled hylafax from the distribution.
> > Ran faxsetup, ran faxmodem, back to abnormal:
> > 
> > Can't use sendfax from command prompt without a password.
> 
> OK. This is a separate problem, methinks. Did you re-run faxsetup?
> Does "faxstat" work? Is your $SPOOL/etc/hosts file for HylaFAX
> permissions set up correctly, which may be the problem?

Yes, I re-ran faxsetup about 5 times, before I decided to reinstall
hylafax from the BSDI 3.0 distribution.  What ARE the permissions supposed
to be for the $SPOOL/etc/hosts file?  I have it set for 0600, both user
and group ownership is "uucp".  One other flexfax subscriber emailed me
and told me that I have to be sure not to have any trailing spaces,

> 
> > The strange thing is that I had this working with the very first install!!
> > What threw me off is that snnp takes the same port as stronghold.  So I
> > reinstalled hylafax without the snnp and moved stronghold to another port.
> 
> Oof!. That's a potetntial problem for the rest of us: thanks for the
> heads up. Is 444 the default port for stronghold?

443 is the SSL port and 444 is the default for the admin portion of
stronghold.  The stronghold tech suggested to move the admin into a
different port, I set it for 446 in the httpd.conf file and then was able
to start stronghold without any problems.
I commented out the snnp port in /etc/services because I don't need it at
this time.


> 
> > When I look at the dir with the faxes that went out previously, they
> > report as being sent as "nobody@myhostname.com".  Which tells me there is
> > some kind of permissions problem here with some of the dirs and files
> > owned by "fax".
> 
> Umm. No. hfaxd sits on the port 4559, and answers requests for service
> from the sendfax or faxstat command.. They correspond, the user's name
> gets associated by hfaxd with the service request (using permissions
> in the hosts file to verify their authority to send faxes) and an
> email address gets generated from that data (as I understand it). The
> data gets fed to the hfaxd by sendfax, and it handles the fax job.  No
> ownership or file permissions involved there.
>

Everytime Ireboot the system, I get an error "socket operation on
non-socket" and hylafax won't always start.  I tried commenting it out of
the /etc/inetd.conf file and start if by hand, but for some reason, when I
reboot, the hfaxd starts up anyway on its own, giving me that error that
it can't bind to port 4559.

There must be something that I am overlooking.

Regards
Eddie

 > Nico Garcia > Engineer, CIRL >Mass. Eye and Ear Infirmary > 			raoul@cirl.meei.harvard.edu
> 
> -----BEGIN PGP SIGNATURE-----
> Version: 2.6.2
> 
> iQCVAwUBNGty1D/+ItycgIJRAQEaWAP/YQ+0m/L9oaiLx6IB/6Az2tTcm3rezDrH
> zz4PTg6OVAle6VrfkwFKRA1gqsLPuUtfk7wB5q8bj14pqRXpoBtvB/CB9q3zAmd7
> WVdxs8sc818l1TOSjA6Huw8goZCFRp6Yxne8mRRr1X6Z+moI8cJkcYdRYalHF82m
> x+056fJ5FsA=
> =/uvk
> -----END PGP SIGNATURE-----
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 14 Nov 1997 11:14:42 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Eddie Reed <eddie@eddiereed.com>
Cc: flexfax@sgi.com
Subject: Re: flexfax: Hylafax - fax user defined in passwd
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Thu, 13 Nov 1997, Eddie Reed wrote:

> > OK. This is a separate problem, methinks. Did you re-run faxsetup?
> > Does "faxstat" work? Is your $SPOOL/etc/hosts file for HylaFAX
> > permissions set up correctly, which may be the problem?
> 
> Yes, I re-ran faxsetup about 5 times, before I decided to reinstall
> hylafax from the BSDI 3.0 distribution.  What ARE the permissions supposed
> to be for the $SPOOL/etc/hosts file?  I have it set for 0600, both user
> and group ownership is "uucp".  One other flexfax subscriber emailed me
> and told me that I have to be sure not to have any trailing spaces,

OK, good. I wasn't sure you had tried those steps: sometimes people
leave stuff out or change things, then are confused when something
they didn't realize was affected gets clobbered.

That "hosts" file should be 0600, uucp owned and uucp group owned.
You could try starting with the hosts file that comes with the
distribution, and seeing if *that* works correctly. Also: how are
you starting hfaxd, starting it with a script at boot time, using
inetd, or what? Do you have only one faxq running and only one hfaxd,
either left running as a boot-time daemon or initiated by inetd?

Faxstat and "telnet localhost 4559" are better tools to see if your
hfaxd is up and running: they don't have to do as many complicated
things, and don't leave failed print jobs in the queue....

> 443 is the SSL port and 444 is the default for the admin portion of
> stronghold.  The stronghold tech suggested to move the admin into a
> different port, I set it for 446 in the httpd.conf file and then was able
> to start stronghold without any problems.
> I commented out the snnp port in /etc/services because I don't need it at
> this time.

OK. I kind of prefer having it there: that way, I can move the
contents of /etc/services around and have HylaFAX follow the changes
(as long as I start hfaxd and any other programs by using the name
"snpp" rather than the number 444).

> Everytime Ireboot the system, I get an error "socket operation on
> non-socket" and hylafax won't always start.  I tried commenting it out of
> the /etc/inetd.conf file and start if by hand, but for some reason, when I
> reboot, the hfaxd starts up anyway on its own, giving me that error that
> it can't bind to port 4559.

*HUH*. This is screwball. OK, definitely take it out of inetd.conf.
Reboot, and see if it is in fact running with a "ps" command.

Then take a look at your boot time initialization scripts. Under
Linux, they live in /etc/rc.d/various sub-directories, and see if
they have anything implanted there by your original installation
to start up the hfaxd.

> There must be something that I am overlooking.

Could be, but it sounds like you're trying the right sort of steps
to find the difficulty.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNGx46z/+ItycgIJRAQGK0AP9HKndnBHP69c9Yuv5zY5RaxbtRs2KcmVn
8s/c4zBKo7WlvM/WkJSDPytdPlD2ejTvfDMECaFL/H0lBjbUuB5jlcF8NLuN5Com
U9ry5/VoYXXk2moCmsSGAunEUB9ZkZ0cFiX8zt+dfNdyCTsxJpo7U+1sBX7MEKn2
z2nkK4S1upA=
=x/4x
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 14 Nov 1997 13:05:37 -0500 (EST)
From: Eddie Reed <eddie@eddiereed.com>
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
cc: flexfax@sgi.com
Subject: Re: flexfax: Hylafax - fax user defined in passwd
Sender: owner-flexfax@celestial.com


> That "hosts" file should be 0600, uucp owned and uucp group owned.
> You could try starting with the hosts file that comes with the
> distribution, and seeing if *that* works correctly. Also: how are

No, already tried:  still asks me for a password.

> you starting hfaxd, starting it with a script at boot time, using
> inetd, or what? 

Yes with inetd by way of the default faxsetup script, but then I comment
it out.

> Do you have only one faxq running and only one hfaxd,
> either left running as a boot-time daemon or initiated by inetd?

yes, 1 each

> 
> Faxstat and "telnet localhost 4559" are better tools to see if your
> hfaxd is up and running: they don't have to do as many complicated
> things, and don't leave failed print jobs in the queue....

 faxstat wants a password too...


> *HUH*. This is screwball. OK, definitely take it out of inetd.conf.
> Reboot, and see if it is in fact running with a "ps" command.

Right: I take it out and it comes back at boottime!

> 
> Then take a look at your boot time initialization scripts. Under
> Linux, they live in /etc/rc.d/various sub-directories, and see if
> they have anything implanted there by your original installation
> to start up the hfaxd.


It starts in rc and is a default set by BSDI.  The script reads that if
/var/spool/fax/etc/config exists, to start faxq and hfaxd with the
parameter -i, start hylafax -o, start oldfax and start snnp.
It further states that out-only should be run from rc.local, but there is
no entry in rc.local.

Before I 'doctor' up some of these files, should I comment the rc section
relating to hfaxd and then just start it up manually with the -i parameter 
once the system has rebooted?

Still not clear though what all that has to do with the password prompt it
insists on giving me.  Another user suggested to insert the following into
the $SPOOL/etc/hosts file:

127.0.0.1:5::

to allow anyone to send a fax without a password prompt.  I figured the
other user assumed my user "fax" to be uid "5", while I have it set at the
default "6" as suggested by the mailing list keeper.

Regards
eddie

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 14 Nov 1997 13:34:26 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Eddie Reed <eddie@eddiereed.com>
Cc: flexfax@sgi.com
Subject: Re: flexfax: Hylafax - fax user defined in passwd
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

We exchanged:

> > Then take a look at your boot time initialization scripts. Under
> > Linux, they live in /etc/rc.d/various sub-directories, and see if
> > they have anything implanted there by your original installation
> > to start up the hfaxd.
> 
> It starts in rc and is a default set by BSDI.  The script reads that if
> /var/spool/fax/etc/config exists, to start faxq and hfaxd with the
> parameter -i, start hylafax -o, start oldfax and start snnp.
> It further states that out-only should be run from rc.local, but there is
> no entry in rc.local.

OK. Got it. What the??? If you're going to start it that way, I
strongly suggest just running the "hylafax.sh" script as part of the
boot process. It's available in the distribution source code if not in
your particular OS's binary distribution. This sets up a single hfaxd
daemon monitoring all 3 ports, and the script can be used to *kill* or
restart the daemons as necessary.

> Before I 'doctor' up some of these files, should I comment the rc section
> relating to hfaxd and then just start it up manually with the -i parameter 
> once the system has rebooted?

This looks like an equally good idea, if not better. This would also let
you start it up with a decent debugger to monitor any difficulties.

> Still not clear though what all that has to do with the password prompt it
> insists on giving me.  Another user suggested to insert the following into
> the $SPOOL/etc/hosts file:
> 
> 127.0.0.1:5::
> 
> to allow anyone to send a fax without a password prompt.  I figured the
> other user assumed my user "fax" to be uid "5", while I have it set at the
> default "6" as suggested by the mailing list keeper.

This may be part of the problem. Also, a plain

	127.0.0.1
	localhost

should be sufficient to run faxstat from the same host. I'm also not
sure if you need to re-start or kick hfaxd after making such changes.



			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNGyZqj/+ItycgIJRAQFqWQQAn2pvli8S6xQOm4+KC6bUVt1dtbRIOAaw
uU1e2ipacEMMwV/mxQus7LidnP6E/qBfFGLEjRNcsEwIrY8ScOnZtGElrBXwDRqw
lpl0Zuqq8p4u+jVd9KSqqWvdSJIfSijEweKrVM1Bp4mJzZ5lpXRFo5kgf217m2nl
j3sr2tDhhYQ=
=Zvbk
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Hylafax - fax user defined in passwd
To: eddie@eddiereed.com (Eddie Reed)
Date: Sat, 15 Nov 1997 08:52:04 +0100 (MEZ)
Cc: raoul@cirl.meei.harvard.edu, flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Eddie Reed wrote:
    
    
    > That "hosts" file should be 0600, uucp owned and uucp group owned.
    > You could try starting with the hosts file that comes with the
    > distribution, and seeing if *that* works correctly. Also: how are

To clarify: HylaFAX's file host(4F) should be owned by user "fax"
which must have the same uid as "uucp". It should also have
mode 0600 (BTW: all these details are really explained in the
man page of HylaFAX's host(4F)).
    
    No, already tried:  still asks me for a password.

This problem is unrelated to the mode bits or owner of host(4F).
A simple test would show you that the message of the server
is "access denied" (and not asking for pw):
# cd spool/etc
# ls -l hosts
-rw-------   1 uucp     uucp         139 Nov  2 17:59 hosts
# chmod 0644 hosts
# telnet localhost hylafax
Trying 127.0.0.1 ...
220 thias server (HylaFAX (tm) Version 4.0pl2beta) ready.
user guru
530 User guru access denied.
quit
221 Goodbye.
Connection closed by foreign host.
# chmod 0600 hosts
# telnet localhost hylafax
Trying 127.0.0.1 ...
220 thias server (HylaFAX (tm) Version 4.0pl2beta) ready.
user guru
230 User guru logged in.
quit
221 Goodbye.
# chown guru hosts
# telnet localhost hylafax
Trying 127.0.0.1 ...
220 thias server (HylaFAX (tm) Version 4.0pl2beta) ready.
user guru
530 User guru access denied.
quit
221 Goodbye.
Connection closed by foreign host.
	...

If the hfaxd(1M) daemon asks for a pw than one reason could be
that it does not use "127.0.0.1" (local loopback) for the incoming
client connection. You may check this also simple by running
a telnet to the port of the hfaxd(1M) and check from another
terminal/session with netstat(1) which interface is used for this
connection, e.g. run
$ netstat -an | grep 4559
tcp        0      0  127.0.0.1.4559         127.0.0.1.4351         ESTABLISHED
tcp        0      0  127.0.0.1.4351         127.0.0.1.4559         ESTABLISHED
tcp        0      0  *.4559                 *.*                    LISTEN
$ 
(assuming the "hylafax" port is 4559)

    > you starting hfaxd, starting it with a script at boot time, using
    > inetd, or what? 
    
    Yes with inetd by way of the default faxsetup script, but then I comment
    it out.

Why do you comment it out? It should work fine (and works for
me fine) from inetd.conf(4F) and also from boot-time script.
    
    > Do you have only one faxq running and only one hfaxd,
    > either left running as a boot-time daemon or initiated by inetd?
    
    yes, 1 each

That's ok.
    
    > 
    > Faxstat and "telnet localhost 4559" are better tools to see if your
    > hfaxd is up and running: they don't have to do as many complicated
    > things, and don't leave failed print jobs in the queue....
    
     faxstat wants a password too...

Naturally. If it asks for a pw while using sendfax(1) it will
also ask for a pw for faxstat(1) (asuming the same environment
for each client).
    
    > *HUH*. This is screwball. OK, definitely take it out of inetd.conf.
    > Reboot, and see if it is in fact running with a "ps" command.
    
    Right: I take it out and it comes back at boottime!

Does your system create a new inetd.conf(4F) file from some
"master file" while booting (some UNIXes are doing such nice things,
e.g. also for the file inittab(4F) -- check your manual of
inetd.conf(4F), inetd(1M) or init(1M) for information about this).
    
    > 
    > Then take a look at your boot time initialization scripts. Under
    > Linux, they live in /etc/rc.d/various sub-directories, and see if
    > they have anything implanted there by your original installation
    > to start up the hfaxd.
    
    
    It starts in rc and is a default set by BSDI.  The script reads that if
    /var/spool/fax/etc/config exists, to start faxq and hfaxd with the
    parameter -i, start hylafax -o, start oldfax and start snnp.
    It further states that out-only should be run from rc.local, but there is
    no entry in rc.local.
    
    Before I 'doctor' up some of these files, should I comment the rc section
    relating to hfaxd and then just start it up manually with the -i parameter 
    once the system has rebooted?

No. This would not solve your problem (hfaxd(1M) is running
and that's fine).
    
    Still not clear though what all that has to do with the password prompt it
    insists on giving me.  Another user suggested to insert the following into
    the $SPOOL/etc/hosts file:
    
    127.0.0.1:5::

This line is only used if hfaxd(1M) uses the "127.0.0.1"
interface for the client connection. See above. The "5" is
also total unrelated to your problem (asking for pw); check
the manual of hosts(4F) for the real purpose of this
parameter.
    
    to allow anyone to send a fax without a password prompt.  I figured the
    other user assumed my user "fax" to be uid "5", while I have it set at the
    default "6" as suggested by the mailing list keeper.

As said allready, the user "fax" must have the same uid as "uucp"
and not 5 or 6 or some other value. Just check your passwd(4F)
file for these users. I *may* be 6 (if "uucp" is 6).

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Sun, 16 Nov 1997 19:30:58 -0500 (EST)
From: Helen Birkmann <hbirkmann@mkl.com>
Reply-To: Helen Birkmann <hbirkmann@mkl.com>
To: Matthias Apitz <Matthias.Apitz@SOFTCON.de>
cc: Eddie Reed <eddie@eddiereed.com>, raoul@cirl.meei.harvard.edu,
        flexfax@sgi.com
Subject: Re: flexfax: Hylafax - fax user defined in passwd
Sender: owner-flexfax@celestial.com

On Sat, 15 Nov 1997 Matthias.Apitz@SOFTCON.de wrote:

> Eddie Reed wrote:
>     
>     
>     > That "hosts" file should be 0600, uucp owned and uucp group owned.
>     > You could try starting with the hosts file that comes with the
>     > distribution, and seeing if *that* works correctly. Also: how are
> 
> To clarify: HylaFAX's file host(4F) should be owned by user "fax"
> which must have the same uid as "uucp". It should also have
> mode 0600 (BTW: all these details are really explained in the
> man page of HylaFAX's host(4F)).

HylaFAX's host file on my system is owned by uucp (both user and group)
and it WILL not let me change it to "fax"  Users fax and uucp both hve
the same uid, and the hosts file DOES have perms 0600.

> Trying 127.0.0.1 ...
> 220 thias server (HylaFAX (tm) Version 4.0pl2beta) ready.
> user guru
> 230 User guru logged in.
> quit
> 221 Goodbye.

It lets me log in as user "anything"!


> If the hfaxd(1M) daemon asks for a pw than one reason could be
> that it does not use "127.0.0.1" (local loopback) for the incoming
> client connection. You may check this also simple by running
> a telnet to the port of the hfaxd(1M) and check from another
> terminal/session with netstat(1) which interface is used for this
> connection, e.g. run
> $ netstat -an | grep 4559


 tcp        0      0  *.4559                 *.*                    LISTEN

this is all I get when running this command


> (assuming the "hylafax" port is 4559)


yes!

> Why do you comment it out? It should work fine (and works for
> me fine) from inetd.conf(4F) and also from boot-time script.

edited both inetd.conf and /etc/rc to start hfaxd with the parameter -I
rather than the default listed in /etc/rc (-i -s -o)

> 
> Does your system create a new inetd.conf(4F) file from some
> "master file" while booting (some UNIXes are doing such nice things,
> e.g. also for the file inittab(4F) -- check your manual of
> inetd.conf(4F), inetd(1M) or init(1M) for information about this).


No, the file is the same after each reboot, because it keeps any changes I
made before reboot.

> No. This would not solve your problem (hfaxd(1M) is running
> and that's fine).

Yes, but the /var/log/messages file shows several errors:

When I keep the hfaxd startup parameter in the inetd.conf, I get an error
"hylafax tcp/ip port 4559 already in use"

When I try to start it manually (commenting the startup parameter in the
inetd.conf) I get the error "non-socket operation on socket", and it won't
start up.

> This line is only used if hfaxd(1M) uses the "127.0.0.1"
> interface for the client connection. See above. The "5" is
> also total unrelated to your problem (asking for pw); check
> the manual of hosts(4F) for the real purpose of this
> parameter.

I started out with a complete new hosts file and made sure there's no
trailing spaces.

> As said allready, the user "fax" must have the same uid as "uucp"
> and not 5 or 6 or some other value. Just check your passwd(4F)
> file for these users. I *may* be 6 (if "uucp" is 6).

What about the daily insecurity output file:


>From root@lumpy.mkl.com Sun Nov 16 19:28:27 1997
Date: Sun, 16 Nov 1997 03:11:03 -0500 (EST)
From: System Administrator <root@lumpy.mkl.com>
To: root@lumpy.mkl.com
Subject: lumpy daily insecurity output


Checking the /etc/master.passwd file:
Login fax is off but still has a valid shell.

/etc/master.passwd has duplicate user id's.
fax 6	uucp 6

Checking home directories:
user fax home directory is owned by uucp

Tx in advance for your advice.

Sincerely
Eddie 

************************************************************************
*	MKL.Net		        	Internet Service Provider      *
*	A service of CPU Industries                                    *
*	Bainbridge, New York, USA      	PPP, dial-up, telnet, uucp,    *
*	http://www.mkl.com		virtual domains, web design    *
*	https://mall.mkl.com	Secured server - Merchant services     *
************************************************************************

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Hylafax - fax user defined in passwd
To: hbirkmann@mkl.com
Date: Mon, 17 Nov 1997 08:27:07 +0100 (MEZ)
Cc: eddie@eddiereed.com (Eddie Reed), raoul@cirl.meei.harvard.edu,
        flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Helen Birkmann wrote:
    
    On Sat, 15 Nov 1997 Matthias.Apitz@SOFTCON.de wrote:
    
    > Eddie Reed wrote:
    >     
    >     
    >     > That "hosts" file should be 0600, uucp owned and uucp group owned.
    >     > You could try starting with the hosts file that comes with the
    >     > distribution, and seeing if *that* works correctly. Also: how are
    > 
    > To clarify: HylaFAX's file host(4F) should be owned by user "fax"
    > which must have the same uid as "uucp". It should also have
    > mode 0600 (BTW: all these details are really explained in the
    > man page of HylaFAX's host(4F)).
    
    HylaFAX's host file on my system is owned by uucp (both user and group)
    and it WILL not let me change it to "fax"  Users fax and uucp both hve
    the same uid, and the hosts file DOES have perms 0600.

If "fax" and "uucp" both have the same uid in the /etc/passwd
it's ok. The uid is stored as a integer value in the inode of
each file and a ls(1) will take this int and uses it to resolve
the name (it looks for a name with that uid) and takes perhaps
the first matching entry.
    
    > Trying 127.0.0.1 ...
    > 220 thias server (HylaFAX (tm) Version 4.0pl2beta) ready.
    > user guru
    > 230 User guru logged in.
    > quit
    > 221 Goodbye.
    
    It lets me log in as user "anything"!

Me too :-))
$ telnet localhost 4559
Trying 127.0.0.1 ...
Connected to localhost.
220 thias server (HylaFAX (tm) Version 4.0pl2beta) ready.
user Birkmann
230 User Birkmann logged in.

And this also normal if your HylaFAX's hosts(4F) file does not
restrict the use of the service from localhost (see the man page
for details).
    
    > If the hfaxd(1M) daemon asks for a pw than one reason could be
    > that it does not use "127.0.0.1" (local loopback) for the incoming
    > client connection. You may check this also simple by running
    > a telnet to the port of the hfaxd(1M) and check from another
    > terminal/session with netstat(1) which interface is used for this
    > connection, e.g. run
    > $ netstat -an | grep 4559
    
    
     tcp        0      0  *.4559                 *.*                    LISTEN
    
    this is all I get when running this command

Did you run the netstat(1) while a telnet session to port
4559 was in progress? If there is no client connection to
port 4559 it will show only the LISTEN, if there is a
connection to port 4559 it will also schow ESTABLISHED lines.
    
    
    > (assuming the "hylafax" port is 4559)
    
    yes!
    
    > Why do you comment it out? It should work fine (and works for
    > me fine) from inetd.conf(4F) and also from boot-time script.
    
    edited both inetd.conf and /etc/rc to start hfaxd with the parameter -I
    rather than the default listed in /etc/rc (-i -s -o)

The "-I" will only work if hfaxd(1M) is started from the inetd(1M)
via an entry in inetd.conf(4F). "-I" means: the client/server
protocol is using stdin and stdout (these file descriptors are
handed over by inetd(1M) as a connected socket). If you use
"-I" in a boot-time start script it will not work. On the
other hand you can't use "-i hylafax" in inetd.conf(4F) because
inetd(1M) is listening on port "hylafax" and not hfaxd(1M) itself.
Your error message below "4559 in use" will be the result of
using "-i hylafax" in inetd.conf(4F).  In inetd.conf(4F) you must
use "-I".
    
    > 
    > Does your system create a new inetd.conf(4F) file from some
    > "master file" while booting (some UNIXes are doing such nice things,
    > e.g. also for the file inittab(4F) -- check your manual of
    > inetd.conf(4F), inetd(1M) or init(1M) for information about this).
    
    
    No, the file is the same after each reboot, because it keeps any changes I
    made before reboot.
    
    > No. This would not solve your problem (hfaxd(1M) is running
    > and that's fine).
    
    Yes, but the /var/log/messages file shows several errors:
    
    When I keep the hfaxd startup parameter in the inetd.conf, I get an error
    "hylafax tcp/ip port 4559 already in use"

Yes. You set "-i hylafax" in inetd.conf(4F) and hfaxd(1M)
sees the port 4559 in use (by inetd(1M). Use "-I".
    
    When I try to start it manually (commenting the startup parameter in the
    inetd.conf) I get the error "non-socket operation on socket", and it won't
    start up.

Please show exactly the command line you're using for the manually
start. Keep also in mind that you have to send the signal
SIG_HUP to inetd(1M) to make your changes to inetd.conf(4F)
working.
    
    > This line is only used if hfaxd(1M) uses the "127.0.0.1"
    > interface for the client connection. See above. The "5" is
    > also total unrelated to your problem (asking for pw); check
    > the manual of hosts(4F) for the real purpose of this
    > parameter.
    
    I started out with a complete new hosts file and made sure there's no
    trailing spaces.
    
    > As said allready, the user "fax" must have the same uid as "uucp"
    > and not 5 or 6 or some other value. Just check your passwd(4F)
    > file for these users. I *may* be 6 (if "uucp" is 6).
    
    What about the daily insecurity output file:
    
    
    >From root@lumpy.mkl.com Sun Nov 16 19:28:27 1997
    Date: Sun, 16 Nov 1997 03:11:03 -0500 (EST)
    From: System Administrator <root@lumpy.mkl.com>
    To: root@lumpy.mkl.com
    Subject: lumpy daily insecurity output
    
    
    Checking the /etc/master.passwd file:
    Login fax is off but still has a valid shell.
    
    /etc/master.passwd has duplicate user id's.
    fax 6	uucp 6
    
    Checking home directories:
    user fax home directory is owned by uucp

I think you should ignore these messages. They're just
telling that the things are fine as they are.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Tue, 18 Nov 1997 09:12:26 -0500 (EST)
From: Helen Birkmann <hbirkmann@mkl.com>
Reply-To: Helen Birkmann <hbirkmann@mkl.com>
To: Matthias Apitz <Matthias.Apitz@SOFTCON.de>
cc: Eddie Reed <eddie@eddiereed.com>, flexfax@sgi.com
Subject: Re: flexfax: Hylafax - fax user defined in passwd
Sender: owner-flexfax@celestial.com

> Did you run the netstat(1) while a telnet session to port
> 4559 was in progress? If there is no client connection to
> port 4559 it will show only the LISTEN, if there is a
> connection to port 4559 it will also schow ESTABLISHED lines.

Yes, I did:

tcp	0	0	127.0.0.1.4559	127.0.0.1.1025	ESTABLISHED
tcp	0	0	127.0.0.1.1025	127.0.0.1.4559	ESTABLISHED
tcp	*.*	4559					LISTEN


> The "-I" will only work if hfaxd(1M) is started from the inetd(1M)
> via an entry in inetd.conf(4F). "-I" means: the client/server
> protocol is using stdin and stdout (these file descriptors are
> handed over by inetd(1M) as a connected socket). If you use
> "-I" in a boot-time start script it will not work. On the
> other hand you can't use "-i hylafax" in inetd.conf(4F) because
> inetd(1M) is listening on port "hylafax" and not hfaxd(1M) itself.
> Your error message below "4559 in use" will be the result of
> using "-i hylafax" in inetd.conf(4F).  In inetd.conf(4F) you must
> use "-I".

inetd.conf:

hylafax  stream  tcp  nowait  fax /usr/contrib/lib/hylafax/hfaxd hfaxd -I

rc:

/usr/contrib/bin/faxq
/usr/contrib/lib/hylafax/hfaxd -i

....

After reboot, only faxq shows up as a process and the error message
during bootup says:
hylafax usage	[-i port] [-o port] [-s port]

I then go to /usr/contrib/lib/hylafax directory and type:

./hfaxd -i 4559

HylaFax starts up and I get the following error in the daemonlog:
HylaFax INET: bind (port 4559) already in use; unable to initialize server

and I get the password prompt when trying to issue sendfax or faxstat
fromt he command line.

REgards
Helen
************************************************************************
*	MKL.Net		        	Internet Service Provider      *
*	A service of CPU Industries                                    *
*	Bainbridge, New York, USA      	PPP, dial-up, telnet, uucp,    *
*	http://www.mkl.com		virtual domains, web design    *
*	https://mall.mkl.com	Secured server - Merchant services     *
************************************************************************

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Hylafax - fax user defined in passwd
To: hbirkmann@mkl.com
Date: Tue, 18 Nov 1997 16:03:50 +0100 (MEZ)
Cc: eddie@eddiereed.com, flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Helen Birkmann wrote:
    
    > Did you run the netstat(1) while a telnet session to port
    > 4559 was in progress? If there is no client connection to
    > port 4559 it will show only the LISTEN, if there is a
    > connection to port 4559 it will also schow ESTABLISHED lines.
    
    Yes, I did:
    
    tcp	0	0	127.0.0.1.4559	127.0.0.1.1025	ESTABLISHED
    tcp	0	0	127.0.0.1.1025	127.0.0.1.4559	ESTABLISHED
    tcp	*.*	4559					LISTEN
    
Yes, and it looks fine now. One established connection from
127.0.0.1.4559 <--> 127.0.0.1.1025 and a listen on all interfaces
for port 4559. Fine.
    
    > The "-I" will only work if hfaxd(1M) is started from the inetd(1M)
    > via an entry in inetd.conf(4F). "-I" means: the client/server
    > protocol is using stdin and stdout (these file descriptors are
    > handed over by inetd(1M) as a connected socket). If you use
    > "-I" in a boot-time start script it will not work. On the
    > other hand you can't use "-i hylafax" in inetd.conf(4F) because
    > inetd(1M) is listening on port "hylafax" and not hfaxd(1M) itself.
    > Your error message below "4559 in use" will be the result of
    > using "-i hylafax" in inetd.conf(4F).  In inetd.conf(4F) you must
    > use "-I".
    
    inetd.conf:
    
    hylafax  stream  tcp  nowait  fax /usr/contrib/lib/hylafax/hfaxd hfaxd -I

The above line looks ok...
    
    rc:
    
    /usr/contrib/bin/faxq
    /usr/contrib/lib/hylafax/hfaxd -i
                             ^^^^^^^^

and this line is wrong (the port is missing after the "-i" flag
and it will also conflict with the entry in inetd.conf(4F). You
should not have both lines. Remove one and re-boot later. For
the moment you can go on because you have a fine listening
daemon for port 4559 (inetd(1M)) and this daemon will launch
a correct hfaxd(1M) for each connection.
    
    ....
    
    After reboot, only faxq shows up as a process

That's also normal because hfaxd(1M) isn't running by itself.
It gets started from the listening daemon inetd(1M) if a connection
on port 4559 comes in.

    and the error message
    during bootup says:
    hylafax usage	[-i port] [-o port] [-s port]

Yes, the port is missing in the start script in rc.
    
    I then go to /usr/contrib/lib/hylafax directory and type:
    
    ./hfaxd -i 4559

Why? You have a working entry in inetd.conf(4F) and the inetd(1M)
is listening on port 4559. Why do you want to start another service
for the same port. Please take a book about TCP/IP, ports and network
services and read also the man page of inetd(1M) to get more
feeling how the things should work.
    
    HylaFax starts up and I get the following error in the daemonlog:
    HylaFax INET: bind (port 4559) already in use; unable to initialize server

Of course the port is in use: inetd(1M) is listening on it.
    
    and I get the password prompt when trying to issue sendfax or faxstat
    fromt he command line.

Ok. But this is unrelated to the above. Please show your entries
in HylaFAX's hosts(4F) file.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Tue, 18 Nov 1997 14:38:52 -0500 (EST)
From: Helen Birkmann <hbirkmann@mkl.com>
To: Matthias Apitz <Matthias.Apitz@SOFTCON.de>
cc: flexfax@sgi.com
Subject: Re: flexfax: Hylafax - fax user defined in passwd
Sender: owner-flexfax@celestial.com

> Ok. But this is unrelated to the above. Please show your entries
> in HylaFAX's hosts(4F) file.

localhost
127.0.0.1

Regards
Helen

> 
> 	matthias
> 

************************************************************************
*	MKL.Net		        	Internet Service Provider      *
*	A service of CPU Industries                                    *
*	Bainbridge, New York, USA      	PPP, dial-up, telnet, uucp,    *
*	http://www.mkl.com		virtual domains, web design    *
*	https://mall.mkl.com	Secured server - Merchant services     *
************************************************************************

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Hylafax - fax user defined in passwd
To: hbirkmann@mkl.com (Helen Birkmann)
Date: Wed, 19 Nov 1997 10:02:19 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Helen Birkmann wrote:
    
    > Ok. But this is unrelated to the above. Please show your entries
    > in HylaFAX's hosts(4F) file.
    
    localhost
    127.0.0.1

The entry "localhost" will allow any user from the local host
who is using the ``localhost'' interface (this 2nd part is also
verry important) access to the facsimile service of HylaFAX.

I've a line in HylaFAX's hosts(4F) file like this:

guru@localhost

Now compare the following two examples:
1.
kant:/home/guru $ faxstat 
HylaFAX scheduler on kant: Running
Modem ttyFN01 (+49.89.61308188): Running and idle
Modem ttyFN02 (+49.89.61308189): Running and idle

2.
kant:/home/guru $ FAXSERVER=kant export FAXSERVER
kant:/home/guru $ uname -n
kant
kant:/home/guru $ faxstat
Password:

In the 1st example the connection uses the ``localhost'' interface
and the line "guru@localhost" in hosts(4F) let me use the
service. In the 2nd example hfaxd(1M) searches for an entry
like "guru@kant" in the hosts(4F) file which isn't there
and it prompts for pw.
As you can see it's also important which interface on the local
host is used for the connection.

Could you please repeat your question? I don't know at the moment
what your problem is (I think the hfaxd(1M) start and port questions
are now resolved).

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
X-Authentication-Warning: tui.pinnacle.co.nz: jonc owned process doing -bs
Date: Fri, 14 Nov 1997 09:53:49 +1300 (NZDT)
From: Jonathan Chen <jonc@pinnacle.co.nz>
To: Helen Birkmann <hbirkmann@mkl.com>
cc: Eddie Reed <eddie@eddiereed.com>, flexfax@sgi.com
Subject: Re: flexfax: permissions
Sender: owner-flexfax@celestial.com

On Thu, 13 Nov 1997, Helen Birkmann wrote:

> On Tue, 11 Nov 1997, Jonathan Chen wrote:
> 
[snip]
> > Specify a specific user-id number for the entries in
> > /var/spool/fax/etc/hosts. Make sure there are no trailing white space
> > or comments.
> 
> In other words, just list users such as 
> www
> 84
> nobody
> 32766
> nogroup
> 32767

Not quite what I meant. A sample /var/spool/fax/etc/hosts file may list:

	127.0.0.1:5::

This will set all faxes sent on the localhost with a fax-userid of 5.
Check out the entry for hosts(4F) for more details.
--
Jonathan Chen <jonc@pinnacle.co.nz>   | "Vini, vidi, velcro...
                                      |    I came, I saw, I stuck around"

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Sat, 08 Nov 1997 11:50:02 -0600
From: Mike Frohme <frohme@memphis.edu>
Organization: Information Systems - University of Memphis
To: flexfax@sgi.com
Subject: flexfax: sendpage+hfaxd, SNPP DATA broken? (4.0pl1 Sol2.5binary)
Sender: owner-flexfax@celestial.com

Hello,

	Just picked up HylaFax 4.0pl1 (Solaris 2.5, binary dist
	from SGI) and have installed.  Fax features work great,
	but I'm having minor problems w/Alpha paging:
	
	sendpage works great when you specify the message in the
	command line.  Specifying the message via STDIN fails - the
	read never completes.

	It appears that the MESS command is AOK, but the DATA command
	hangs in this distribution.  

telnet localhost 444
Trying 127.0.0.1...
Connected to localhost.
Escape character is '^]'.
220 palantir.memphis.edu SNPP server (HylaFAX (tm) Version 4.0pl1)
ready.
login XX
230 User XX logged in.
page frohme
250 Pager ID accepted; provider: 9,XXXXXXX pin: XXXXXXX jobid: 35.
DATA
354 Begin message input; end with <CRLF>'.'<CRLF>.
Test message
Line #2
.

	No amount of cajoling, pleading, encoding, etc seems to get the read
	to finish - aborting the telnet session is the only way out.  When
	done via sendpage with the -v option, the only difference seems
	to be that the system knows the data is there (the file is 134
	bytes long)...

%> sendpage -v -p frohme < week

[snipped]

250 Pager ID accepted; provider: XXXXXXX pin: XXXXXXX jobid: 36.
destination pin frohme: request id is 36 for host localhost
SEND message data, 134 bytes
-> DATA
354 Begin message input; end with <CRLF>'.'<CRLF>.
-> .

(^C to abort)

	The jobs never appear in the queue, but the job id does
	increment. Any known issues surrounding this behavior?

	I've tried without much luck to get a SessionTrace on
	this problem - no log file ever appears in logs/. Not
	sure what I'm missing there.
	
	Thanks in advance for any help/suggestions you might
	have - even for getting a detailed log for isolating the
	behavior.

-- 
Mike Frohme                                     frohme@memphis.edu
Information Systems - The University of Memphis       901/678-8820

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Comments: Authenticated sender is <spine001@gold.tc.umn.edu>
From: Julio C Spinelli <spine001@gold.tc.umn.edu>
To: flexfax@sgi.com, Camden Clarke <cclarke@inconceivable.com>
Date: Thu, 13 Nov 1997 12:26:12 +0000
Subject: flexfax: Re: egetty 
CC: glotel@protocom.com
Sender: owner-flexfax@celestial.com

Sorry that it took so long for me to follow on on this subject, but I 
have been very busy at work.

After thorough investigation I found that the problem in Linux 2.0.29 
is that when Getty:run does a dup2 to duplicate the modem file 
descriptor into the standard output, in the child process, this destroys the file 
descriptor that point to the modem in the parent process. Why does 
this happens, it escapes me totally. I may look into the kernel 
source during my vacations (may be not...)

Whenever the dup2 is executed in the child the parent process can not 
read or write to the modem when the child exits.

The only way I managed to make it work for Linux is to have the 
egetty have uid 0 and open and close the modem itself. By doing this 
the only change needed in Getty.c++ is an additional if statement 
that does not do any dup2 is done when egetty is called.

Matthias, are you interested in this for the pl2?
> 
> You don't actually have to open the modem in your egetty program; HylaFAX
> will leave you with stdin and stdout setup to the modem.  This is intentional,
> so that egetty doesn't have to know where to find the modem, or deal
> with opening it correctly for shared access (e.g. Suns).
>                 -Cam Clarke
>  
> In astatos.flex-fax you write:
>  
> >Julio C Spinelli wrote:
> >
> >    Thanks for the info. egetty is nothing, it does not exist at all. The
> >    only meaning of the string egetty is that if EGettyArgs is defined as
> >    a non nul string and AnswerRotary is set to "extern". Hylafax will
> >    call a program /bin/egetty. If that program does not exist...
> >
> >    It is the user/system developer's responsibility to write that
> >    program. This program's job is to answer the incoming call, determine
> >    its type and then exit with a CALLTYPE_XXX (FAX, DATA, DONE, ERROR,
> >    VOICE) code. Hylafax will then proceed as if it itself would have
> >    determined the type of call.
> >
> >    The issue I have is that when I do this I get a modem write error
> >    after calling the sys::write(fd,......   call and as I don't fully
> >    understand the inheritance rules that Sam applies and neither have I
> >    been able to understand the Dispatch class. I am at a loss in trying
> >    to guess what is going wrong.
> >
> >    This is why I am looking for any working example for simple and bare
> >    bones it is.
> >
> >    Thanks for your help,
> >
> >    Julio
> >
> >    Note: I hope this clarifies what egetty is.
>  
> >Try this small example. It works for me (don't blame me about
> >this C-code, it's not the way I'm writing real code :-))
> >It works for me and faxgetty(1M) receives the fax fine (see the log).
>  
> >   matthias
>  
>  
> >/* change this define to the real device on your system */
> >#define LINE "/dev/ttyFN02"
> >main()
> >{
> >   int fd;
> >   fd = open(LINE, 2);
> >   if(fd < 0)
> >       exit(0);
> >   write(fd, "ATA\r", 4);   /* time to pick up the call */
> >   close(fd);
> >   exit(2);
> >}
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: per-user cover pages?
To: ben@dha.unon.org (Ben Parker)
Date: Thu, 13 Nov 1997 14:54:21 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Ben Parker wrote:
    
    
    Can't find anything on this in the archives. Is there a way to let
    different users define their own cover page?

Read the man pages for the env variable FAXCOVER;

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 13 Nov 1997 10:06:59 -0500 (EST)
From: Evan Leibovitch <evan@telly.org>
To: Ben Parker <ben@dha.unon.org>
cc: flexfax@sgi.com
Subject: Re: flexfax: per-user cover pages?
Sender: owner-flexfax@celestial.com


On Thu, 13 Nov 1997, Ben Parker wrote:

> Can't find anything on this in the archives. Is there a way to let
> different users define their own cover page?

They key to this is the manpage for 'faxcover', which is the part of the
process that generates a complete cover page from a template and adds
transmission-specific information.

faxcover allows you to specify a different template by checking the
FAXCOVER environment variable. So you can have different template
locations specified in users' .profile settings by setting FAXCOVER
appropriately. 

------------------------------------------------------------------------------
 Evan Leibovitch, Sound Software Ltd, located in beautiful Brampton, Ontario
Supporting PC-based Unix since 1985 / Caldera & SCO authorized / www.telly.org
----------------- HURD is to Linux as Plan 9 is to System V ------------------

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 13 Nov 1997 11:31:30 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Ben Parker <ben@dha.unon.org>
Cc: flexfax@sgi.com
Subject: Re: flexfax: per-user cover pages?
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Thu, 13 Nov 1997, Ben Parker wrote:

> Can't find anything on this in the archives. Is there a way to let
> different users define their own cover page?

"man sendfax", and "man faxcover". Each user will need to call the
commands and specifically select their coverpages. Alternatively, you
can setup a script like "sfax" or tools like "faxtool" to assign the
users a per-user selected cover page, if available. Or, you may be
able to set it up on a per-user basis in the "~/.faxdb" file, which
affects the behavior of faxcover.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNGsrXD/+ItycgIJRAQGJFAP/ejKEW9yW85ibsriIyhQF+eWqmSikeCN7
YzO1GT6UMvpAlPf6VIK3Ivwnymk7t8AGSTzozYxG/Z+vwmq1/GRd5V3k146CZJrH
wrfe+El97Z0dOxeOEhZnAeDogRygRZwtZw6WyMRhoYlL3dbFEab/nsuAEY9k5l80
EPJ1LQdPelo=
=StoV
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 13 Nov 1997 12:58:48 -0600
From: Dan Freise <dan@unt.edu>
Organization: UNT
To: flexfax@sgi.com
Subject: flexfax: DTMF routing using Hylafax??
Sender: owner-flexfax@celestial.com

I've seen a couple of questions in the arcives about DTMF routing, but I
didn't see any answers...

The man pages (at least the ones I've read) don't mention it, but does
anyone know if it is possible to do DTMF routing of incoming faxes using
a Zyxel 1496 modem?

-- 
Dan Freise                 ____  http://autobahn.cascss.unt.edu/~dan
A&S Computing Services     \  /
(817) 565-4498              \/   dan@unt.edu

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 13 Nov 1997 20:09:43 +0100
From: Ramon Arxer <arxer@bcn.servicom.es>
Reply-To: arxer@bcn.servicom.es
Organization: http://yi.com/home/ArxerRamon/
To: "flexfax@sgi.com" <flexfax@sgi.com>
Subject: flexfax: sendfax again...
Sender: owner-flexfax@celestial.com

Hello!

First of all I've to thank everybody that answered my last message in
this list.
After reading the compillation of the messages in the HylaFAX archive,
I'm fighting again the sendfax program. It stops when it's communicating
with the HylaFAX server.

When I execute the following command:
	sendfax -n -d 7050 -vv < ~/msg

I get:

Trying localhost (127.0.0.1) at port 4559...
Connected to localhost.
220 gea04.pangea.org server (HylaFAX (tm) Version 4.0pl1) ready.
-> USER root
[...]
213 CHOPTHRESHOLD set to 3.
-> JPARM DOCUMENT /tmp/doc56.ps
200 Added document /tmp/doc56.ps as docq/doc56.ps.78.
-> JSUBM

and sendfax stops, had someone been in the same situation?
May be, it's a problem with the FaxQueuer?

-- 
Thank you very much,

Ramon Arxer i Galabert


e-mail:
	arxer@bcn.servicom.es
	ramon25@casal.upc.es
WWW:
	http://yi.com/home/ArxerRamon/
	http://casal.upc.es/~ramon25/

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: Re: egetty
To: spine001@gold.tc.umn.edu (Julio C Spinelli)
Date: Thu, 13 Nov 1997 20:52:39 +0000 (GMT)
Cc: flexfax@sgi.com, cclarke@inconceivable.com, glotel@protocom.com
Sender: owner-flexfax@celestial.com

> After thorough investigation I found that the problem in Linux 2.0.29 
> is that when Getty:run does a dup2 to duplicate the modem file 

The patches between 2.0.29 and 2.0.31 for tty_io.c amount to 730 lines
(diff -u); I think upgrading the kernel may be your first port of call
(all the patches are in the 2.0.31 update).  2.0.31 was current as of
last weekend.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Re: egetty
To: spine001@gold.tc.umn.edu (Julio C Spinelli)
Date: Fri, 14 Nov 1997 09:18:42 +0100 (MEZ)
Cc: flexfax@sgi.com, cclarke@inconceivable.com, glotel@protocom.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Julio C Spinelli wrote:
    
    Sorry that it took so long for me to follow on on this subject, but I 
    have been very busy at work.
    
    After thorough investigation I found that the problem in Linux 2.0.29 
    is that when Getty:run does a dup2 to duplicate the modem file 
    descriptor into the standard output, in the child process, this destroys the file 
    descriptor that point to the modem in the parent process. Why does 
    this happens, it escapes me totally. I may look into the kernel 
    source during my vacations (may be not...)
    
    Whenever the dup2 is executed in the child the parent process can not 
    read or write to the modem when the child exits.

Try to separate this problem from HylaFAX by writing a small
C example (without using HylaFAX at all) and post a problem
report about this to the linux developers news groups.
    
    The only way I managed to make it work for Linux is to have the 
    egetty have uid 0 and open and close the modem itself. By doing this 
    the only change needed in Getty.c++ is an additional if statement 
    that does not do any dup2 is done when egetty is called.
    
    Matthias, are you interested in this for the pl2?

Without knowledge about the problem in the Linux kernel I'm
unsure to apply a patch like skipping the dup(2) and let
egetty re-open the modem again. I'll try to find the time
to reproduce this on the Linux box of my son.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Re: HylaFAX
To: juvvadi@lekha.org (Ramana Juvvadi)
Date: Fri, 14 Nov 1997 09:36:17 +0100 (MEZ)
Cc: Matthias.Apitz@SOFTCON.de, juvvadi@lekha.org, flexfax@sgi.com,
        raoul@cirl.meei.harvard.edu
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Ramana Juvvadi wrote:
    
    Matthias.Apitz@SOFTCON.de writes:
    
     > Letting the thing go on with an invalid uid for "fax" will lead
     > to problems in HylaFAX which most of the Linux users don't understand
     > and could not resolve. The bottom line of this fault (not running
    
    I am not advocating people not to run faxsetup. They are supposed 
    to run faxsetup after installing the rpm. The rpm adds the user fax
    so that most people don't have any problems during faxsetup.
    As you know the faxsetup linux script for adding a user is broken.

Until now I only tested *modifying* an existing user "fax"
with the faxsetup(1M) script and there is a bug in the Linux's
moduser(1M) command (returning 12 in case the HOME allready
exist). If there is a problem in *adding* a user too, please let
me know the details or post a fix (e.g. a change for the script
sbin/faxsetup.linux).

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 14 Nov 1997 08:58:48 -0500
From: Aleksandar Simic <ASimic@compuserve.com>
Subject: flexfax: 550: /tmp/doc5.ps: Not Owner
To: "flexfax@sgi.com" <flexfax@sgi.com>
Sender: owner-flexfax@celestial.com

Help help help please.

I'm having problems sending faxes. Error message is:

        550: /tmp/doc5.ps Not Owner


and the log file says:

......HylaFAX[28200]: /tmp/doc5.ps: chown: Invalid argument
......HylaFAX[28200]: /sendq/q5: chown: Invalid argument

Here is my environment:

        HP-UX 10.01 running on HP9000 S800 K100
        gcc-2.7.2.2
        HylaFAX-v4.pL1

I did search through archives, troubleshooting section etc. but no luck.
My /var/spool/fax/etc/hosts file conatains:

        127.0.0.1:5::

as suggested earlier by Jonathan Chen. Still no luck.
While going through archives I realized that some people earlier had
similar problems.
Is it that permissions I've setup are wrong ? Or something else.

Any help is greatly appreciated.

Aleksandar Simic (asimic@4compsys.com, asimic@compuserve.com)

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 14 Nov 1997 10:08:16 -0600 (CST)
From: Bill Suetholz <bill@centonline.com>
To: Julio C Spinelli <spine001@gold.tc.umn.edu>
cc: "Mr. Arlington Hewes" <tpcadmin@info.tpc.int>,
        Chad M Vande Pol <chad@is6.iserv.net>, flexfax@sgi.com
Subject: Re: flexfax: Multiple Modems 
Sender: owner-flexfax@celestial.com

Well, Digi has something that could be made to work for their PortServer
line of products.  It is in the unsupported area (use at your own risk).
It is called rtty.  This program uses a PTY port and maps it to a direct
connection to an individual port on the Portserver.  A couple of problems
that I ran into in my situation;  1) I'm on AT&T Unix SVR3  and had
troubles getting it to work with it's networking and PTY's,  2) The darn
thing keeps the port assigned to one machine all the time.  I only want
the connection grabbed when needed so other machines can use it.

Anyways it's there for Sun, SCO, HP, and maybe Linux (haven't looked
recently). www.dgii.com, or ftp.dgii.com:/pub/shareware/unix/rtty

Bill Suetholz 

>On Fri, 7 Nov 1997, Julio C Spinelli wrote:
>
> Humm...,
> 
> It seems to me that what we would need to do is to write a driver 
> that maps a device to a socket and then run hylafax with the name of 
> that device as the serial device.
> 
> Can any Unix guru please indicate if this is the way to go?
> 
> If it is, it would be a question of finding a group interested in 
> developing this. I'm willing to help, but I lack the necessary 
> expertise,
> 
> Julio
> > To:            "Prof. Julio C. Spinelli" <spine001@gold.tc.umn.edu>
> > Cc:            Chad M Vande Pol <chad@is6.iserv.net>,
> >                flexfax@sgi.com
> > Subject:       Re: flexfax: Multiple Modems 
> > Date:          Fri, 07 Nov 1997 12:02:28 +0000
> > From:          "Mr. Arlington Hewes" <tpcadmin@info.tpc.int>
> 
> > >>>>> On Thu, 6 Nov 1997, "PJCS" == Prof. Julio C. Spinelli wrote:
> > 
> >   PJCS> Could you please illuminate us about these modems? Are they class 2
> >   PJCS> 2.0 or class 1 modems? Can you provide some reference to a web page
> >   PJCS> or document that describes their functions/commands.
> > 
> > They are class 2 aware rockwell modems. For information on ASCEND products, 
> > please see www.ascend.com
> > 
> >   PJCS> I don't know if everybody else is on the same boat, but I never
> >   PJCS> heard of them (may be a shame).
> > 
> > Many people will have heard of them . . . mainly ISPs and people in 
> > industrial situations.
> > 
> >   PJCS> If they are of a standard class, hylafax is so versatile that we
> >   PJCS> should be able to manage them,
> > 
> > Well, the problem arises in the need to be able to manage them over TCP/IP, 
> > as they are not directly connected to any machine you see.
> > 
> > -DPN
> > 
> > -- 
> > Mr. Arlington Hewes           (tpcadmin@info.tpc.int)
> > The TPC.INT Subdomain          (http://www.tpc.int/)
> > 
> >        **************************************************
> >        ***  FOR GENERAL INFORMATION                   ***
> >        ***         Mail to tpcfaq@info.tpc.int        ***
> >        ***  FOR A LIST OF CURRENT COVERAGE            ***
> >        ***         Mail to tpccover@info.tpc.int      ***
> >        **************************************************
> > 
> > 
> > 
> > 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 14 Nov 1997 11:18:38 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: Ramon Arxer <arxer@bcn.servicom.es>
Cc: "flexfax@sgi.com" <flexfax@sgi.com>
Subject: Re: flexfax: sendfax again...
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Thu, 13 Nov 1997, Ramon Arxer wrote:

> First of all I've to thank everybody that answered my last message in
> this list.
> After reading the compillation of the messages in the HylaFAX archive,
> I'm fighting again the sendfax program. It stops when it's communicating
> with the HylaFAX server.
> 
> When I execute the following command:
> 	sendfax -n -d 7050 -vv < ~/msg

Well, for one thing, you don't need the "<". Sendfax should expect
filenames of text, TIFF, or Postscript files at this point.

faxstat works right?

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNGx51z/+ItycgIJRAQG+RwP/dfJq/LzvnCiNQliPuIXaTdF/58/tfKL/
rEownPxemLbv6ZLwxjQQFQh0EiyxmbHI+wEv7wfOD1s5LSMTGNFBb5UuEEmzyK8Q
QwQ/d2PPqlILNmA7qVGR86QHbeHYTdgoIgHnQf+X0PXWJqiGro3OcCIlciz1m9Iv
/2ADsGjwhAI=
=05kw
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 14 Nov 1997 19:31:56 +0100
From: Ramon Arxer i Galabert <arxer@bcn.servicom.es>
Organization: http://yi.com/home/ArxerRamon/
To: Nico Garcia <raoul@cirl.meei.harvard.edu>,
        Llista distribucio HylaFAX <flexfax@sgi.com>
Subject: flexfax: sendfax again... (II)
Sender: owner-flexfax@celestial.com

Maybe I should explain a little more of my system.

When I execute the following command:
       sendfax -n -d 7050 -vv  ~/msg

sendfax stops, 

* faxstat, gives the following answer:
     HylaFAX scheduler on gea04.pangea.org: Running

     Modem cua1 (+34.3.401.71.82): Running and idle


* the message log file contains:

Nov 11 18:49:39 gea04 HylaFAX[154]: USER root
Nov 11 18:49:39 gea04 HylaFAX[154]: TYPE I
Nov 11 18:49:39 gea04 HylaFAX[154]: PORT 195.77.46.49;4100
Nov 11 18:49:39 gea04 HylaFAX[154]: MODE Z
Nov 11 18:49:39 gea04 HylaFAX[154]: STOT
Nov 11 18:49:39 gea04 HylaFAX[154]: Filesystem has SysV-style file
creation semantics.

* comunication between sendfax and hylafax:

Trying localhost (127.0.0.1) at port 4559...
Connected to localhost.
220 gea04.pangea.org server (HylaFAX (tm) Version 4.0pl1) ready.
-> USER root
230 User root logged in.
[...]
-> JPARM CHOPTHRESHOLD 3
213 CHOPTHRESHOLD set to 3.
-> JPARM DOCUMENT /tmp/doc56.ps
200 Added document /tmp/doc56.ps as docq/doc56.ps.78.
-> JSUBM

It seems that sendfax stops when trying to submit the job, but I don't
know why.!
Has anybody ever seen something like this?
Thank you very much.
-- 
Best regards,

Ramon Arxer i Galabert

e-mail :
	arxer@bcn.servicom.es
	ramon25@casal.upc.es
WWW :
	http://yi.com/home/ArxerRamon
	http://casal.upc.es/~ramon25

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 14 Nov 1997 12:17:04 -0800 (PST)
From: Tim Rice <tim@trr.metro.NET>
To: Aleksandar Simic <ASimic@compuserve.com>
cc: "flexfax@sgi.com" <flexfax@sgi.com>
Subject: Re: flexfax: 550: /tmp/doc5.ps: Not Owner
Sender: owner-flexfax@celestial.com

On Fri, 14 Nov 1997, Aleksandar Simic wrote:

> Help help help please.
> 
> I'm having problems sending faxes. Error message is:
> 
>         550: /tmp/doc5.ps Not Owner
> 

I don't run HP-UX, but I tracked down the same problem on SCO OS5
and we fixed it in pl1.

Check your port.h for CONFIG_MAXGID. If it is set to 60002, find
out what went wrong with configure.
THe relevent lines are.

    # Select the maximum fax UID.  This value is stored
    # in the GID of files so must be constrained by any
    # system limitations.  The maximum value is used for
    # anonymous logins.  The default value of 60002 comes
    # from NFS/nobody usage on various systems.
    #
    test "$CONFIG_MAXGID" || {
	case $TARGET in
	*-sco*)		CONFIG_MAXGID=60000;;
	*-hpux*)	CONFIG_MAXGID=60000;;
	*)		CONFIG_MAXGID=60002;;	# default
	esac
    }
> 
> and the log file says:
> 
> ......HylaFAX[28200]: /tmp/doc5.ps: chown: Invalid argument
> ......HylaFAX[28200]: /sendq/q5: chown: Invalid argument
> 
> Here is my environment:
> 
>         HP-UX 10.01 running on HP9000 S800 K100
>         gcc-2.7.2.2
>         HylaFAX-v4.pL1
> 
> I did search through archives, troubleshooting section etc. but no luck.
> My /var/spool/fax/etc/hosts file conatains:
> 
>         127.0.0.1:5::
> 
> as suggested earlier by Jonathan Chen. Still no luck.
> While going through archives I realized that some people earlier had
> similar problems.
> Is it that permissions I've setup are wrong ? Or something else.
> 
> Any help is greatly appreciated.
> 
> Aleksandar Simic (asimic@4compsys.com, asimic@compuserve.com)
> 

--
Tim Rice			Multitalents	(707) 887-1469 (voice)
tim@trr.metro.net

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Sat, 15 Nov 1997 08:19:48 +0100
From: Martin Tanzer <office@webpower.co.at>
Organization: ArtCom
To: flexfax@sgi.com
Subject: flexfax: Need help on Linux
Sender: owner-flexfax@celestial.com

Hi, I am trying to set up Hylafax  4.0p11 on two Linux boxes  (as root)
with the same result: access denied. I searched in the list and did
everything I found posted:

The fax and the uucp have the same UID,  the same group (uucp), even the
same password. The /var/spool/fax/etc/hosts belongs to uucp and has the
correct mode (600). When I start hfaxd I get the message in the message
file : HylaFAX INET: bind (port 4559): Address already in use, unable to
init server.
When I kill faxd and faxq I still have no access to the ports with
faxstat -v or telnet localhost 4559.

I've seen that Linix and Hylafax is a bit tricky - I hope you can help.
I spent lots of hours with this stuff.

Martin

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Need help on Linux
To: office@webpower.co.at (Martin Tanzer)
Date: Sat, 15 Nov 1997 09:18:24 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Martin Tanzer wrote:
    
    Hi, I am trying to set up Hylafax  4.0p11 on two Linux boxes  (as root)
    with the same result: access denied. I searched in the list and did
    everything I found posted:
    
    The fax and the uucp have the same UID,  the same group (uucp), even the
    same password. The /var/spool/fax/etc/hosts belongs to uucp and has the
    correct mode (600). When I start hfaxd I get the message in the message
    file : HylaFAX INET: bind (port 4559): Address already in use, unable to
    init server.

Check with netstat(1) if there is a LISTEN for this port
and if so try to find out which daemon (inetd(1M)?) is
listening on port 4559. If you have a line for this service
in your file inetd.conf(4F) check the parameters against
the man page of hfaxd(1M). If you use startup from inetd.conf(4F)
you must use the flag "-I" and not "-i hylafax".

    When I kill faxd and faxq I still have no access to the ports with
    faxstat -v or telnet localhost 4559.

There is no faxd(1M) in HylaFAX v4.0pl1 (and if you have faxd you
have a mix of versions on your system) and faxq(1M) has nothing
to do with the port 4559.
    
    I've seen that Linix and Hylafax is a bit tricky - I hope you can help.
    I spent lots of hours with this stuff.

No. Not Linux && HylaFAX are tricky. I get more and more the feeling
(esp. since I've had a look onto the Linux of my son) that Linux
comes with some (faulty) HylaFAX or FlexFAX or mgetty+sendfax
installations or preparations and also some bugs/problems in the 
binary packages provided by the CD-ROM supplier. Just remove
everything from your system which smells like "fax" (also from
the sysem files like passwd(4F) and inetd.conf(4F), pull the
source (v4.0pl1), configure, make and install from scratch
(and run faxsetup(1M) :-)).

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Comments: Authenticated sender is <spine001@gold.tc.umn.edu>
From: Julio C Spinelli <spine001@gold.tc.umn.edu>
To: Matthias.Apitz@SOFTCON.de
Date: Sat, 15 Nov 1997 10:42:05 +0000
Subject: Re: flexfax: Need help on Linux
CC: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

I can testify that Matthias ia absolutely right, I have Linux 2.0.29 
(in the process of upgrading to 2.0.31, big changes in the tty 
access) and downloaded the hylafax source, did configure, make, make 
install and faxsetup and wow! it worked with no problem. With the 
huge advantage that I can change the source and recompile if needed!

Only one point, if you don't want to have problems, pick a modem that 
has been mentioned to work reliably in the list and for which a 
config file has been posted and copy that file on top of the one that 
faxaddmodem creates for that modem. (please still run faxaddmodem).

One more comment, make sure that you either answer yes to the 
question on whether you should place the hfaxd daemon in inet.conf or 
that if your machine will have a lot of client traffic, that you add 
by hand the hfaxd in one of your init files (check tha man page for 
details on the options).

Then either start the faxgetty cuax (x=0,1,2,3 etc.) by hand or add 
it to your init file.

By the way my linux distribution is slackware,

Best luck,

Julio

Note: Mattias, Is there a list of the major bugs in hylafax4.pl1 
posted anywhere? Thanks.

> From:          Matthias.Apitz@SOFTCON.de
> Subject:       Re: flexfax: Need help on Linux
> To:            office@webpower.co.at (Martin Tanzer)
> Date:          Sat, 15 Nov 1997 09:18:24 +0100 (MEZ)
> Cc:            flexfax@sgi.com
> Reply-to:      Matthias.Apitz@SOFTCON.de (Matthias Apitz)

> Martin Tanzer wrote:
>     
>     Hi, I am trying to set up Hylafax  4.0p11 on two Linux boxes  (as root)
>     with the same result: access denied. I searched in the list and did
>     everything I found posted:
>     
>     The fax and the uucp have the same UID,  the same group (uucp), even the
>     same password. The /var/spool/fax/etc/hosts belongs to uucp and has the
>     correct mode (600). When I start hfaxd I get the message in the message
>     file : HylaFAX INET: bind (port 4559): Address already in use, unable to
>     init server.
> 
> Check with netstat(1) if there is a LISTEN for this port
> and if so try to find out which daemon (inetd(1M)?) is
> listening on port 4559. If you have a line for this service
> in your file inetd.conf(4F) check the parameters against
> the man page of hfaxd(1M). If you use startup from inetd.conf(4F)
> you must use the flag "-I" and not "-i hylafax".
> 
>     When I kill faxd and faxq I still have no access to the ports with
>     faxstat -v or telnet localhost 4559.
> 
> There is no faxd(1M) in HylaFAX v4.0pl1 (and if you have faxd you
> have a mix of versions on your system) and faxq(1M) has nothing
> to do with the port 4559.
>     
>     I've seen that Linix and Hylafax is a bit tricky - I hope you can help.
>     I spent lots of hours with this stuff.
> 
> No. Not Linux && HylaFAX are tricky. I get more and more the feeling
> (esp. since I've had a look onto the Linux of my son) that Linux
> comes with some (faulty) HylaFAX or FlexFAX or mgetty+sendfax
> installations or preparations and also some bugs/problems in the 
> binary packages provided by the CD-ROM supplier. Just remove
> everything from your system which smells like "fax" (also from
> the sysem files like passwd(4F) and inetd.conf(4F), pull the
> source (v4.0pl1), configure, make and install from scratch
> (and run faxsetup(1M) :-)).
> 
> 	matthias
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Need help on Linux
To: spine001@gold.tc.umn.edu (Julio C Spinelli)
Date: Sun, 16 Nov 1997 07:45:17 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Julio C Spinelli wrote:
    
    	...
    
    Note: Mattias, Is there a list of the major bugs in hylafax4.pl1 
    posted anywhere? Thanks.

I host a list of all real bug reports (together with
the relevant information and/or patches). There is also
a TODO list in the source directory.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Need help on Linux
To: martin@webpower.co.at
Date: Mon, 17 Nov 1997 07:55:04 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Martin Tanzer wrote:
    
    Thanks for your quick answer, Mathias. I followed your advice and pulled
    the code and started from the scratch. I didn't remove the old version,
    I just overwrote it. I restarted Hylafax and  I am able to telnet port
    4559 and to run faxstat -v without 'access denied' error. I still get an
    error message with (Login failed: 550 cannot set privileges), but I am
    sure I am a BIG step further. I think you should know that I use SuSE
    5.0. MAybe this saves other users lots of time.
    
    Martin

I quick glance to the code in hfaxd/ source directory shows:
$ fgrep -i "cannot set privileges" *.c++
Login.c++:      reply(550, "Cannot set privileges.");
$ 

The reason for this reply in Login.c++ is a failing chroot(2)
or chdir(2) call at line ~150 in Login.c++. Check why the hfaxd(1M)
is unable to perform these calls. Maybe it's not started by root
or installed without setuid bit turned on (e.g. mode 4755) or the
root of the spooling area is a symlink to some other place.
BTW: One could check with strace(1) which of the above sys calls
is failing.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 17 Nov 1997 11:31:19 +0100 (MET)
From: Carsten Hoeger <choeger@suse.de>
To: Matthias.Apitz@SOFTCON.de, Matthias Apitz <guru@sisis.de>
cc: Martin Tanzer <office@webpower.co.at>, flexfax@sgi.com
Subject: Re: flexfax: Need help on Linux
Sender: owner-flexfax@celestial.com

On Sat, 15 Nov 1997 Matthias.Apitz@SOFTCON.de wrote:

> Martin Tanzer wrote:
>     
>     Hi, I am trying to set up Hylafax  4.0p11 on two Linux boxes  (as root)
>     with the same result: access denied. I searched in the list and did
>     everything I found posted:
>     
>     The fax and the uucp have the same UID,  the same group (uucp), even the
>     same password. The /var/spool/fax/etc/hosts belongs to uucp and has the
>     correct mode (600). When I start hfaxd I get the message in the message
>     file : HylaFAX INET: bind (port 4559): Address already in use, unable to
>     init server.
> 
> Check with netstat(1) if there is a LISTEN for this port
> and if so try to find out which daemon (inetd(1M)?) is
> listening on port 4559. If you have a line for this service
> in your file inetd.conf(4F) check the parameters against
> the man page of hfaxd(1M). If you use startup from inetd.conf(4F)
> you must use the flag "-I" and not "-i hylafax".
> 
>     When I kill faxd and faxq I still have no access to the ports with
>     faxstat -v or telnet localhost 4559.
> 
> There is no faxd(1M) in HylaFAX v4.0pl1 (and if you have faxd you
> have a mix of versions on your system) and faxq(1M) has nothing
> to do with the port 4559.
>     
>     I've seen that Linix and Hylafax is a bit tricky - I hope you can help.
>     I spent lots of hours with this stuff.
> 
> No. Not Linux && HylaFAX are tricky. I get more and more the feeling
> (esp. since I've had a look onto the Linux of my son) that Linux
> comes with some (faulty) HylaFAX or FlexFAX or mgetty+sendfax
> installations or preparations and also some bugs/problems in the 
> binary packages provided by the CD-ROM supplier. Just remove
> everything from your system which smells like "fax" (also from
> the sysem files like passwd(4F) and inetd.conf(4F), pull the
> source (v4.0pl1), configure, make and install from scratch
> (and run faxsetup(1M) :-)).

Let me point to our Support Database, espacially for Martin Tanzer.

The following URL should help german reading users -this articles will be
available at our American site www.suse.com, soon-:


http://www.suse.de/Support/sdb/choeger_hylafax-problems.html

Mr. Tanzer didn't read /usr/doc/packages/hylafax/README.SuSE, that says:


[...]
During the execution of `faxsetup` you have to answer the question:

        Should an entry be added to /etc/inetd.conf [no]?

with no because we start HylaFAX(tm) standalone via init-script.
[...]

So he has an entry in /etc/inetd.conf, wich should be removed.


mfG,
	Carsten Hoeger
------
Carsten Hoeger  - S.u.S.E. GmbH -  Gebhardtstr. 2  -  90762 Fuerth  -  Germany
fax +49-911-3206727      web http://www.suse.de        mailbox +49-911-3247122
"Ein Buch ist eine Wolke, auf der man keinen Husten bekommt." (Max Goldt)

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 17 Nov 1997 11:32:56 +0100 (MET)
From: Carsten Hoeger <choeger@suse.de>
To: Julio C Spinelli <spine001@gold.tc.umn.edu>
cc: Matthias.Apitz@SOFTCON.de, flexfax@sgi.com
Subject: Re: flexfax: Need help on Linux
Sender: owner-flexfax@celestial.com

On Sat, 15 Nov 1997, Julio C Spinelli wrote:

[...]

> Then either start the faxgetty cuax (x=0,1,2,3 etc.) by hand or add 
> it to your init file.

By the way: Don't use one of the /dev/cuax devices. Use /dev/ttySx
instead...


mfG,
	Carsten Hoeger
------
Carsten Hoeger  - S.u.S.E. GmbH -  Gebhardtstr. 2  -  90762 Fuerth  -  Germany
fax +49-911-3206727      web http://www.suse.de        mailbox +49-911-3247122
"Ein Buch ist eine Wolke, auf der man keinen Husten bekommt." (Max Goldt)

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 17 Nov 1997 11:36:05 +0100 (MET)
From: Carsten Hoeger <choeger@suse.de>
To: Martin Tanzer <martin@webpower.co.at>
cc: Matthias Apitz <Matthias.Apitz@SOFTCON.de>, flexfax@sgi.com
Subject: Re: flexfax: Need help on Linux
Sender: owner-flexfax@celestial.com

On Mon, 17 Nov 1997, Martin Tanzer wrote:

> Thanks for your quick answer, Mathias. I followed your advice and pulled
> the code and started from the scratch. I didn't remove the old version,
> I just overwrote it. I restarted Hylafax and  I am able to telnet port
> 4559 and to run faxstat -v without 'access denied' error. I still get an
> error message with (Login failed: 550 cannot set privileges), but I am
> sure I am a BIG step further. I think you should know that I use SuSE
> 5.0. MAybe this saves other users lots of time.

Lots of time may be saved, if users read the given documentation...

Sorry, but I must say that... :-(

mfG,
	Carsten Hoeger
------
Carsten Hoeger  - S.u.S.E. GmbH -  Gebhardtstr. 2  -  90762 Fuerth  -  Germany
fax +49-911-3206727      web http://www.suse.de        mailbox +49-911-3247122
"Ein Buch ist eine Wolke, auf der man keinen Husten bekommt." (Max Goldt)

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Need help on Linux
To: choeger@suse.de (Carsten Hoeger)
Date: Mon, 17 Nov 1997 11:57:13 +0100 (MEZ)
Cc: spine001@gold.tc.umn.edu (Julio C Spinelli), flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Carsten Hoeger wrote:
    
    On Sat, 15 Nov 1997, Julio C Spinelli wrote:
    
    [...]
    
    > Then either start the faxgetty cuax (x=0,1,2,3 etc.) by hand or add 
    > it to your init file.
    
    By the way: Don't use one of the /dev/cuax devices. Use /dev/ttySx
    instead...

Could you please explain in detail the difference between
the two devices (as far I can see they have at least different
major number and perhaps different kernel drivers):

$ ls -l /dev/ttyS1 /dev/cua1
crw-rw-rw-   1 root     uucp       5,  65 Jul 18  1994 /dev/cua1
crw-rw-rw-   1 root     tty        4,  65 Jul 18  1994 /dev/ttyS1

Why /dev/cuax exists _and_ should not be used?

	matthias

-- 
firm: matthias.apitz@sisis.de [voc:+49-89-61308-351, fax: +49-89-61308-188]
priv: guru@thias.muc.de
 PGP: Key fingerprint = 0C 01 F2 23 EC 17 A2 D5  46 2D 29 4C 0E 8B 7E 8F

from USENET:
People who run servers understand that flashy interactive interfaces have
nothing to do with the underlying functionality and often get in the way.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 17 Nov 1997 14:23:22 +0100 (MET)
From: Carsten Hoeger <choeger@suse.de>
To: Matthias Apitz <Matthias.Apitz@SOFTCON.de>
cc: Julio C Spinelli <spine001@gold.tc.umn.edu>, flexfax@sgi.com
Subject: Re: flexfax: Need help on Linux
Sender: owner-flexfax@celestial.com

On Mon, 17 Nov 1997 Matthias.Apitz@SOFTCON.de wrote:

> Carsten Hoeger wrote:
>     
>     On Sat, 15 Nov 1997, Julio C Spinelli wrote:
>     
>     [...]
>     
>     > Then either start the faxgetty cuax (x=0,1,2,3 etc.) by hand or add 
>     > it to your init file.
>     
>     By the way: Don't use one of the /dev/cuax devices. Use /dev/ttySx
>     instead...
> 
> Could you please explain in detail the difference between
> the two devices (as far I can see they have at least different
> major number and perhaps different kernel drivers):
> 
> $ ls -l /dev/ttyS1 /dev/cua1
> crw-rw-rw-   1 root     uucp       5,  65 Jul 18  1994 /dev/cua1
> crw-rw-rw-   1 root     tty        4,  65 Jul 18  1994 /dev/ttyS1
> 
> Why /dev/cuax exists _and_ should not be used?


I don't know exactly why, but you can for example read in the PPP-HOWTO:


[...]

  9.2.  Serial Port Names

  Historically, Linux used cuaX devices for dial out and ttySx devices
  for dial in.

  The kernel code that required this was changed in kernel version 2.0.x
  and you should now use ttySx for both dial in and dial out. I
  understand that the cuaX device names may well disappear in future
  kernel versions.

[...]

mfG,
	Carsten Hoeger
------
Carsten Hoeger  - S.u.S.E. GmbH -  Gebhardtstr. 2  -  90762 Fuerth  -  Germany
fax +49-911-3206727      web http://www.suse.de        mailbox +49-911-3247122
"Ein Buch ist eine Wolke, auf der man keinen Husten bekommt." (Max Goldt)

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Leslie Mikesell <les@Mcs.Net>
Subject: Re: flexfax: Need help on Linux
To: Matthias.Apitz@SOFTCON.de
Date: Mon, 17 Nov 1997 10:20:19 -0600 (CST)
Cc: choeger@suse.de, spine001@gold.tc.umn.edu, flexfax@sgi.com
Sender: owner-flexfax@celestial.com

>     By the way: Don't use one of the /dev/cuax devices. Use /dev/ttySx
>     instead...
> 
> Could you please explain in detail the difference between
> the two devices (as far I can see they have at least different
> major number and perhaps different kernel drivers):
> 
> $ ls -l /dev/ttyS1 /dev/cua1
> crw-rw-rw-   1 root     uucp       5,  65 Jul 18  1994 /dev/cua1
> crw-rw-rw-   1 root     tty        4,  65 Jul 18  1994 /dev/ttyS1
> 
> Why /dev/cuax exists _and_ should not be used?

Ancient history: Back in the bad old days before smartmodems, there
really was a separate device used to place outgoing calls.  Then
even after smartmodems were developed that could take 'in band'
dialing commands on the same serial port, uucp and cu (the stock
unix dialup suite) had no mechanism to open a port that did not
have carrier detect active.  If you configured the modem to hold
CD up all the time, you couldn't tell when the other end hung up.
Also, there were locking issues between getty and the dial-out
programs if you wanted bidirectional lines.  One of the early
solutions to this was to revive the concept of the cuaxx devices
but instead of a separate dialer device they became a different
interface to the same serial port as a ttyxx line.  The cuaxx
allowed opening without CD so you could chat with a modem and the
kernel provided locking between the two interfaces so a getty
waiting for CD to complete the open on the ttyxx device would not
wake up if another process had opened the matching cuaxx device for
an outbound call.  This was a workable scheme, but it screws up
the lockfile concepts used by most software and it is no longer
necessary.  More recent versions of unix provide ways to open the
ttyxx devices without CD up and the uucp/cu setup now has a way
to describe it.  Programs expecting to block waiting for CD on
inbound calls have to use the ttyxx devices, and if you are going
to mix these with programs that use lockfiles to contend for
outbound access you have to use the same device there as well.

  Les Mikesell
     les@mcs.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 17 Nov 1997 12:20:57 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
Cc: Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: Need help on Linux
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Mon, 17 Nov 1997, Carsten Hoeger wrote:

> On Mon, 17 Nov 1997 Matthias.Apitz@SOFTCON.de wrote:
> 
> > Carsten Hoeger wrote:
> >     
> >     On Sat, 15 Nov 1997, Julio C Spinelli wrote:
> >     
> >     [...]
> >     
> >     > Then either start the faxgetty cuax (x=0,1,2,3 etc.) by hand or add 
> >     > it to your init file.
> >     
> >     By the way: Don't use one of the /dev/cuax devices. Use /dev/ttySx
> >     instead...
> > 
> > Could you please explain in detail the difference between
> > the two devices (as far I can see they have at least different
> > major number and perhaps different kernel drivers):
> > 
> > $ ls -l /dev/ttyS1 /dev/cua1
> > crw-rw-rw-   1 root     uucp       5,  65 Jul 18  1994 /dev/cua1
> > crw-rw-rw-   1 root     tty        4,  65 Jul 18  1994 /dev/ttyS1
> > 
> > Why /dev/cuax exists _and_ should not be used?
> 
> 
> I don't know exactly why, but you can for example read in the PPP-HOWTO:

It's a legacy of when some OS'es, such as SunOS, used one device for
incoming calls and another for outgoing calls. This unfortunately made
using the same lock files rather awkward, and could confuse the hell
out of systems like HylaFAX that have a single process (faxgetty)
monitoring the line and spinning off getty's, fax process, etc. as
necessary.

According to Ted T'so at MIT, who maintains the Linux archives, it's
being written out of future releases: it's much too difficult and
confusing to maintain.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNHB87z/+ItycgIJRAQG9PQP8C53EDfoQ6jODON3P2Hl3b/+6j0BpQRL4
7M9d4phZi727TDcjryqR4+hnI0anikqUrcv/nUryA0lhE79lX0GLkifjMy8D7txv
FFd2W9NvKDuSDT1wFdQ1gOI6GHUlliZw4cPY4tUdHaNh+jjnBMtiHOu+OTb0oY16
9ySZRAuZ2Pw=
=PqHb
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 17 Nov 1997 14:19:06 -0500 (EST)
From: Evan Leibovitch <evan@telly.org>
To: Carsten Hoeger <choeger@suse.de>
cc: Matthias Apitz <Matthias.Apitz@SOFTCON.de>,
        Julio C Spinelli <spine001@gold.tc.umn.edu>, flexfax@sgi.com
Subject: Re: flexfax: Need help on Linux
Sender: owner-flexfax@celestial.com


>   Historically, Linux used cuaX devices for dial out and ttySx devices
>   for dial in.
> 
>   The kernel code that required this was changed in kernel version 2.0.x
>   and you should now use ttySx for both dial in and dial out. I
>   understand that the cuaX device names may well disappear in future
>   kernel versions.

But you can't count on all the serial-board vendors doing the same thing.
Some contunue to have different operation for cu* and tty*.

------------------------------------------------------------------------------
 Evan Leibovitch, Sound Software Ltd, located in beautiful Brampton, Ontario
Supporting PC-based Unix since 1985 / Caldera & SCO authorized / www.telly.org
----------------- HURD is to Linux as Plan 9 is to System V ------------------

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Comments: Authenticated sender is <spine001@gold.tc.umn.edu>
From: Julio C Spinelli <spine001@gold.tc.umn.edu>
To: Matthias.Apitz@SOFTCON.de
Date: Mon, 17 Nov 1997 17:10:58 +0000
Subject: Re: flexfax: Need help on Linux
CC: Julio C Spinelli <spine001@gold.tc.umn.edu> (Julio C Spinelli),
        flexfax@sgi.com
Sender: owner-flexfax@celestial.com

What I know about the subject is that ttySx are the input devices and 
cuax are the output devices, hence for years people have adopted the 
trick of using cuax when doing input output, like in hylafax. I've 
seen ttySx used mostly when you are going to leave a listening 
program connected to the port only, like in uugetty.

I've also seen an email in this list where they mention that the 
split input output tirck stinks and that the MIT guys are going to 
discontinue it in the next release of LInux.

I've been using cuax for some time with no problem, but....

I hope this helps,

Julio
> Reply-to:      Matthias.Apitz@SOFTCON.de (Matthias Apitz)

> Carsten Hoeger wrote:
>     
>     On Sat, 15 Nov 1997, Julio C Spinelli wrote:
>     
>     [...]
>     
>     > Then either start the faxgetty cuax (x=0,1,2,3 etc.) by hand or add 
>     > it to your init file.
>     
>     By the way: Don't use one of the /dev/cuax devices. Use /dev/ttySx
>     instead...
> 
> Could you please explain in detail the difference between
> the two devices (as far I can see they have at least different
> major number and perhaps different kernel drivers):
> 
> $ ls -l /dev/ttyS1 /dev/cua1
> crw-rw-rw-   1 root     uucp       5,  65 Jul 18  1994 /dev/cua1
> crw-rw-rw-   1 root     tty        4,  65 Jul 18  1994 /dev/ttyS1
> 
> Why /dev/cuax exists _and_ should not be used?
> 
> 	matthias
> 
> -- 
> firm: matthias.apitz@sisis.de [voc:+49-89-61308-351, fax: +49-89-61308-188]
> priv: guru@thias.muc.de
>  PGP: Key fingerprint = 0C 01 F2 23 EC 17 A2 D5  46 2D 29 4C 0E 8B 7E 8F
> 
> from USENET:
> People who run servers understand that flashy interactive interfaces have
> nothing to do with the underlying functionality and often get in the way.
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 17 Nov 1997 15:24:01 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
Cc: Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: Need help on Linux
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Mon, 17 Nov 1997, Evan Leibovitch wrote:

> >   Historically, Linux used cuaX devices for dial out and ttySx devices
> >   for dial in.
> > 
> >   The kernel code that required this was changed in kernel version 2.0.x
> >   and you should now use ttySx for both dial in and dial out. I
> >   understand that the cuaX device names may well disappear in future
> >   kernel versions.
> 
> But you can't count on all the serial-board vendors doing the same thing.
> Some contunue to have different operation for cu* and tty*.

If they are writing Linux compatible drivers, they had darn well
better include th default tty* behavior. And that is *sufficient* for
HylaFAX and modern modem software. Even if they include funky different
options for cu*. Thus, the *availability* of cu* ports becomes moot.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNHCn1j/+ItycgIJRAQGHCQP+IU/JnZ9yRl/CeFlJItKEs/9yq+Trmv7I
C0wfAVdHgGQmj6E8dGYLCpVt5rXfgIf6YjfcXpvMjTtdti5LKaraGP8iJ79mvZld
Qvwa5UP470tsnno7pp+fWS6gLWX7ZEyYcRb6gWOdxoaxec6CAtMJ/32H71ozkbqo
of1bZ7PrZLA=
=kZS+
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Tue, 18 Nov 1997 09:48:47 +0100 (MET)
From: Carsten Hoeger <choeger@suse.de>
To: Evan Leibovitch <evan@telly.org>
cc: Matthias Apitz <Matthias.Apitz@SOFTCON.de>,
        Julio C Spinelli <spine001@gold.tc.umn.edu>, flexfax@sgi.com
Subject: Re: flexfax: Need help on Linux
Sender: owner-flexfax@celestial.com

On Mon, 17 Nov 1997, Evan Leibovitch wrote:

> 
> >   Historically, Linux used cuaX devices for dial out and ttySx devices
> >   for dial in.
> > 
> >   The kernel code that required this was changed in kernel version 2.0.x
> >   and you should now use ttySx for both dial in and dial out. I
> >   understand that the cuaX device names may well disappear in future
> >   kernel versions.
> 
> But you can't count on all the serial-board vendors doing the same thing.
> Some contunue to have different operation for cu* and tty*.

Can you explain, what you mean?
What have different serial-boards from different serial-board vendors to
do with the Linux kernel's dialin/dialout feature?

mfG,
	Carsten Hoeger
------
Carsten Hoeger  - S.u.S.E. GmbH -  Gebhardtstr. 2  -  90762 Fuerth  -  Germany
fax +49-911-3206727      web http://www.suse.de        mailbox +49-911-3247122
"Ein Buch ist eine Wolke, auf der man keinen Husten bekommt." (Max Goldt)

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Tue, 18 Nov 1997 11:04:24 -0500 (EST)
From: Evan Leibovitch <evan@telly.org>
To: Carsten Hoeger <choeger@suse.de>
cc: Matthias Apitz <Matthias.Apitz@SOFTCON.de>,
        Julio C Spinelli <spine001@gold.tc.umn.edu>, flexfax@sgi.com
Subject: Re: flexfax: Need help on Linux
Sender: owner-flexfax@celestial.com


On Tue, 18 Nov 1997, Carsten Hoeger wrote:

> On Mon, 17 Nov 1997, Evan Leibovitch wrote:
  
> > >   Historically, Linux used cuaX devices for dial out and ttySx devices
> > >   for dial in.
 
> > >   The kernel code that required this was changed in kernel version 2.0.x
> > >   and you should now use ttySx for both dial in and dial out. I
> > >   understand that the cuaX device names may well disappear in future
> > >   kernel versions.
 
> > But you can't count on all the serial-board vendors doing the same thing.
> > Some contunue to have different operation for cu* and tty*.
> 
> Can you explain, what you mean?
> What have different serial-boards from different serial-board vendors to
> do with the Linux kernel's dialin/dialout feature?

The vendors of serial-port adaptors may not necessarily use the same
driver conventions -- or even the same utilities -- as the conventional
Linux serial driver. Comtrol Rocketports, for instance, recommend the
'setrocket' program rather than 'setserial'.

The various hardware makers are in control of their own drivers, and
most of the continue to make drivers that have non-modem-control ports
(/dev/cur?? for Comtrol, /dev/cuq[1-4][a-d]?? for Equinox, /dev/cuf??
for Computone) as well as their modem-control tty* equivalents.

On a Computone, when configuring HylaFax, /dev/cuf0 works fine for dialout
but /dev/ttyF0 does not. All I am saying is that one cannot expect or
assume that the behavior of third-party serial boards (especially
'intelligent' ones) will exactly copy that of COM1 and COM2. 

------------------------------------------------------------------------------
 Evan Leibovitch, Sound Software Ltd, located in beautiful Brampton, Ontario
Supporting PC-based Unix since 1985 / Caldera & SCO authorized / www.telly.org
----------------- HURD is to Linux as Plan 9 is to System V ------------------

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Sat, 15 Nov 1997 15:11:11 -0500
From: Yves Trudeau <ytrudeau@jyrmatec.com>
Reply-To: ytrudeau@jyrmatec.com
Organization: Jyrmatec Inc.
To: flexfax@sgi.com
Subject: Re: flexfax: aborting faxes
Sender: owner-flexfax@celestial.com

Nico Garcia wrote:
> 
> -----BEGIN PGP SIGNED MESSAGE-----
> 
> On Wed, 12 Nov 1997, Luiz Otavio L. Zorzella wrote:
> 
> > Hi,
> >
> > I'm trying to share my modem between hylafax and my pppd.
> >
> > I'm being able to do it using the "faxstate" command: I issue a
> > "faxstate -s busy" before I call pppd, so that hylafax won't mess with
> > my connection, and a "faxstate -s ready" after the pppd goes down, so
> > that the waiting faxes will be sent.
> >
> > My problem is that I'm trying to make this thing automatic (putting
> > this commands in my scripts), but I'm having problems when the pppd
> > trys to "go up", and hylafax is using the modem.
> 
> This should be unnecessary. HylaFAX and uucp and kermit and tip and
> minicom and pppd should all be looking in the same directory for the
> same style of lock files: this is /var/lock on Linux, /var/spool/locks
> on SunOS, etc.
> 
> Try running each program and seeing where and what kind of lock file it
> generates: these really should interoperate.
> 
>                         Nico Garcia

I have a similar problem to. I use v4.0pl1 on Linux 2.0.29. Everythings
works fine when hylafax is idle but if there is jobs in the sendq it's a
different story. I tried "faxstate cua1 -s down", "TimeOfDay:  
0500-0530" in config.cua1 (I need the modem between 5:00 and 5:00 AM)
and faxquit with no success. If faxgetty is no running on cua1, faxq
will in fact exit after the "faxquit" command... only after all the jobs
in sendq are done. We do a lot of broadcast faxing so there is always
jobs in sendq. I have already hack faxq in order to make it generate a
"faxq.pid" files which a can use to kill it but this means that the
current jobs will not be notified. Any input is greatly welcomed.

Yves Trudeau, Ph. D.
Jyrmatec Inc.
1381, Jolliet
Sherbrooke, Quebec
Canada, J1K 1Y8
Tel.:(819) 565-8349
Fax.:(819) 565-8273
email: ytrudeau@jyrmatec.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: aborting faxes
To: ytrudeau@jyrmatec.com
Date: Sun, 16 Nov 1997 08:35:18 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Yves Trudeau wrote:
    
    Nico Garcia wrote:
    > 
    > -----BEGIN PGP SIGNED MESSAGE-----
    > 
    > On Wed, 12 Nov 1997, Luiz Otavio L. Zorzella wrote:
    > 
    > > Hi,
    > >
    > > I'm trying to share my modem between hylafax and my pppd.
    > >
    > > I'm being able to do it using the "faxstate" command: I issue a
    > > "faxstate -s busy" before I call pppd, so that hylafax won't mess with
    > > my connection, and a "faxstate -s ready" after the pppd goes down, so
    > > that the waiting faxes will be sent.
    > >
    > > My problem is that I'm trying to make this thing automatic (putting
    > > this commands in my scripts), but I'm having problems when the pppd
    > > trys to "go up", and hylafax is using the modem.
    > 
    > This should be unnecessary. HylaFAX and uucp and kermit and tip and
    > minicom and pppd should all be looking in the same directory for the
    > same style of lock files: this is /var/lock on Linux, /var/spool/locks
    > on SunOS, etc.
    > 
    > Try running each program and seeing where and what kind of lock file it
    > generates: these really should interoperate.
    > 
    >                         Nico Garcia
    
    I have a similar problem to. I use v4.0pl1 on Linux 2.0.29. Everythings
    works fine when hylafax is idle but if there is jobs in the sendq it's a
    different story.

What's the story? If there are jobs in the sendq faxq(1M) will
process them if the modem is free. If the modem is free or not
will be detected by lock-files. The consequence of this it that
all software tools sharing the modem *must* use the same
locking style (e.g. the same filename and the same format
of information in this file). Filenames are usually derived from
information about the device (e.g. major and minor number
of the modem dev) or from the name of the device itself. If you
use different drivers (using different major/minor numbers) for
the same physical device or use different device names it will
cause problems. This logic is simple and clear.
Just watch with the ls(1) and cat(1) or hd(1)
command which lock-file HylaFAX's daemons are using (just call
the line from a telephone while faxgetty(1M) is answering and
go to the lock directory and use ls(1)) and also watch which
lock-file your other software is using.

    I tried "faxstate cua1 -s down", "TimeOfDay:  
    0500-0530" in config.cua1 (I need the modem between 5:00 and 5:00 AM)

TimeOfDay: belongs to the server config file config(4F) and not
to the modem config file. The timerange you set is the time
frame for sending outbound jobs. If you don't like outbound
traffic from 5 AM to 5:30 AM you have to set 0530-0500.
Check the man page for more details.

Normally faxstate(1M) sends the message to the faxgetty(1M)
process.  If there is no faxgetty(1M) running you have to use
the flag "-n" to send the message directly to the faxq(1M).

    and faxquit with no success. If faxgetty is no running on cua1, faxq
    will in fact exit after the "faxquit" command... only after all the jobs
    in sendq are done. We do a lot of broadcast faxing so there is always
    jobs in sendq. I have already hack faxq in order to make it generate a
    "faxq.pid" files which a can use to kill it but this means that the
    current jobs will not be notified. Any input is greatly welcomed.

faxquit(1M) should quit faxq(1M). The stop isn't immediately if
there are active jobs (e.g. jobs for which faxq(1M) has spawned a
child proc like faxsend(1M) or imaging). I've just tested it:

# bin/faxstat -s
HylaFAX scheduler on thias: Running
Modem ttyFN01 (+49.89.xxxxxxxx): Waiting for login session to terminate

JID  Pri S  Owner Number       Pages Dials     TTS Status
32   127 W   guru 12345678      0:0   0:12         
# sbin/faxquit
# bin/faxstat -s
HylaFAX scheduler on thias: Not running
Modem ttyFN01 (+49.89.xxxxxxxx): Waiting for login session to terminate

JID  Pri S  Owner Number       Pages Dials     TTS Status
32   127 W   guru 12345678      0:0   0:12         
#

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 17 Nov 1997 14:44:15 -0500
From: Yves Trudeau <ytrudeau@jyrmatec.com>
Reply-To: ytrudeau@jyrmatec.com
Organization: Jyrmatec Inc.
To: flexfax@sgi.com
Subject: Re: flexfax: aborting faxes
Sender: owner-flexfax@celestial.com

Nico Garcia wrote:
> 
> On Sat, 15 Nov 1997, Yves Trudeau wrote:
> 
> > I have a similar problem to. I use v4.0pl1 on Linux 2.0.29. Everythings
> > works fine when hylafax is idle but if there is jobs in the sendq it's a
> > different story. I tried "faxstate cua1 -s down", "TimeOfDay:
> > 0500-0530" in config.cua1 (I need the modem between 5:00 and 5:00 AM)
> > and faxquit with no success. If faxgetty is no running on cua1, faxq
> > will in fact exit after the "faxquit" command... only after all the jobs
> > in sendq are done. We do a lot of broadcast faxing so there is always
> > jobs in sendq. I have already hack faxq in order to make it generate a
> > "faxq.pid" files which a can use to kill it but this means that the
> > current jobs will not be notified. Any input is greatly welcomed.
> 
> Hmm. Are you using a faxgetty entry in your inittab? And I am confused
> that these programs should *not* respect the HylaFAX lock state, and
> vice versa. What exactly do you need to use as well as HylaFAX, and
> where is it putting the lock files? Remember: just because they are on
> the Linux CD or web sites does not mean they are configured
> identically.
> 
>                         Nico Garcia
>                         Engineer, CIRL
>                         Mass. Eye and Ear Infirmary
>                         raoul@cirl.meei.harvard.edu

Following the suggestion of Matthias, I put a TimeOfDay restriction in
etc/config and it is effective. The only drawback is that we have many
fax lines connected to this server and... they all stop sending faxes
during the restriction. The other program I want to use is pppd. The
lock files are OK, these programs never attemp to use the modem at the
same time. The problem is that I want Hylafax to release one of the
modem at a given time for pppd, and, when there are jobs in sendq (we
have a large clients database, 10000+), I found that it is quite hard to
do this. In the last few days I tried a lot of things:

Without faxgetty running on cua1

#faxq

OK, faxq starts and there is no faxes going out

#faxstate -s ready -n cua1

OK, cua1 starts sending faxes

#faxstate -s down -n cua1

Absolutely no effect... (after 10 minutes, still sending faxes)
Let's try something else,

#faxquit cua1
FIFO.cua1: open: No such device or address

I suppose this is due to the absence of faxgetty?
Another try!

#faxquit 

OK, faxq is dead.

Now, I tried the same with 2 modems sending faxes and... 

#faxq
#faxstate -s ready -n cua1
#faxstate -s ready -n cua2
#faxquit

Now, faxquit is ineffective. The only way to stop now is to kill faxq or
using faxconfig TimeOfDay: "0000-0001". I think (from what I have seen
and understand from the source file faxQueueApp.c++) that faxq wait
until there is no active job before exiting. This is a good behavior as
long as is does not continue to dial new jobs. With more than one modem,
there is almost always a job going on one of the modems. Is it possible
that faxq does not check if quit is true before making a job active. As
you can see... I am not an expert in c++.

The story is similar with faxgetty running (execpt for the FIFO error).

Thank you in advance,

-- 

Yves Trudeau, Ph. D.
Jyrmatec Inc.
1381, Jolliet
Sherbrooke, Qubec
Canada, J1K 1Y8
Tel.:(819) 565-8349
Fax.:(819) 565-8273
email: ytrudeau@jyrmatec.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: flexfax: Re: Linux - SuSE 5.0 - How I,ve got Hylafax working
To: office@webpower.co.at (Martin Tanzer)
Date: Tue, 18 Nov 1997 17:43:10 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Martin Tanzer wrote:
    
    There is a bug in SuSE 5.0 distribution: When you run faxsetup ther are
    I couple of directories missing.

The HylaFAX installation process goes like that:

1. pull and unpack the source
2. run "./configure"
   this step tailors the HylaFAX source for your system into
   at least two global points:
   - it checks your build environment for tools (like compiler,
     TIFF-libs,...) and checks for functions, commands and flags
     to emulate;
   - lets you deside where the files should be placed (e.g. where
     $SPOOLDIR is);
3. run "make"
   this compiles and links the binaries
4. run "make install"
   this creates needed directories, installs the binaries
   and other files with the appr. rights and links
5. run "sh faxsetup" in the server bin dir
   this step checks the runtime environment of HylaFAX
6. use it

Ok, fine. What are "binary distributions" (like the above one)?
The purpose of a binary distribution is to let the user not
run the steps 1-4 (e.g. because he/she has no cc or other needed
tools) and start with something like step 4 (which places all
files to the places where they *must* be placed). The "must" is
really important because the builder of the binaries has done
step 2 and 3 and now the "knowledge" about where the things should
go are in the various a.out's and scripts. Installing them
on some other place as configured in step 2 will break most
of them. That's it.

I don't know in detail the "binary distribution" you have but
if you have a binary distribution which doesn't come along
with some tools and scripts... (like pkgadd on SVR4 or RPM or ...)
which install the files where they have to be installed together
with the appr. rights you should just trash this.

	matthias

    I created these files, and faxsetup
    reported no errors. But the directories belonged to root.root and not to
    uucp.uucp. (I just forgot later, that I created these directories, so I
    didn't think of checking there owner). So I got 'access denied' -
    messages trying to telenet prt 4559 or to run faxstat -v.
    
    I changed the ownership to uucp.uucp:
    
    cd /var/spool/fax
    mkdir archive client dev docq doneq info log pollq recvq sendq status
    tmp
    chown uucp.uucp *
    chmod 700 archive docq doneq pollq sendq tmp
    
    After that Hylafax worked.
    
    
    Thanks to Mathias and Carsten from SuSE, Martin

From root@gateway.cre8tivegroup.com  Wed Feb 17 13:55:48 1999
Sender: root@gateway.cre8tivegroup.com
Date: Wed, 19 Nov 1997 09:29:07 -0500 (EST)
Organization: The Creative Group
From: Patrick Gardella <patrick@cre8tivegroup.com>
To: Matthias.Apitz@SOFTCON.de
Subject: Re: status of the patches
Cc: (Mailinglist for TestGroup HylaFAX) <hylafax-testers@SOFTCON.de>

I've got a *feature* for pl2.

I didn't get a answer from the list on this, so I would guess its a bug.
_______________

I'm running 4.0pl1 on FreeBSD 2.2.1 (Pentium 200 128Megs RAM) with 8 Multitech
ZDK2834 Multimodems.  

HylaFAX is running just fine.  I've got it configured as an email to fax 
gateway.  However, I'm trying to get LineWrap to work right.

from man faxmail:
       LineWrap       Control whether long text lines are wrapper
                      or truncated  at  the  right  hand  margin.
                      (Equivalent to the -c option.)

       Tag               Type         Default         Description
       LineWrap          boolean      Yes            wrap/truncate text lines

Now if I set LineWrap:  Yes in /usr/local/lib/fax/hyla.conf, I get the lines
truncated.  No problem there.

If I set it to No, the lines go off the end of the page with no wrap whatsoever.
So I lose whatever didn't fit on the page.  Ok...

If I don't have LineWrap in hyla.conf at all, the lines wrap, at the right edge
of the fax, but they get cut off in the middle of the words.  It looks like at
80 characters, it cuts it off, whether or not there is a word there or not. 
Isn't it supposed to cut it at around 80 characters at word boundaries? 

My hyla.conf is:
KillTime:       "now +3 days"
MaxDials:       30
GaudyHeaders:   No
Headers:        clear
PageSize:       na-let
TextPointSize:  14pt
TextFont:       Times-Roman

Any suggestions?  Searching the mla for "wrap" "truncate" or "-c" didn't yeild
any answers.

Patrick

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Wed, 7 Jan 1998 11:26:05 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
To: "swamilux@del2" <swamilux@del2.vsnl.net.in>
Cc: flexfax@sgi.com
Subject: Re: flexfax: hylaFAX
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Wed, 7 Jan 1998, swamilux@del2 wrote:

> Dear Sir,
> 
> Your name has been referred to us by ADRIAN SUNDERLAND,Graffin Interne
> Fax regarding setting up of Fax server, Telephony, and other operating
> systems.
> 
> In this connection, we want to intimate you about the prospects of
> INTERNET related services in India.

[ A long and fascinating letter about the need for a commercial fax
server system in India deleted for brevity. ]

> There are so many 'Internet' equipment suppliers' mushrooming all over
> India, and with our intrastructure we can carve over own market, in
> India.  In this connection, we visited your 'website' but we are not
> clear about the various options enumerated therein.
> 
> We shall be pleased to receive your reply in this connection about our
> proposal be impart of 'fax servers' from your side.
> 
> Thanking you,
> 
> Yours faithfully,
> 
> K N SWAMY

Oh, my. As one of the readers and beta testers for the HylaFAX system,
I think I can speak for the rest of us in saying that we are flattered.
Since HylaFAX is not commercial software, but rather user-supported
"freeware", perhaps I can answer some of your concerns.

HylaFAX is written mostly by Sam Leffler of SGI, and is being
maintained by Matthias Appitz who monitors the flexfax@sgi.com mailing
list. Sam wrote it and Matthias maintains it out of an interest in the
technical problems and a desire for good fax software for UNIX
systems.

It is free to download from the FTP site and use: the documentation
online at locations like http://www.vix.com/hylafax is what there
is. New technical problems or software needs are solved by writing
them yourself or asking on the flexfax@sgi.com mailing list if someone
has seen a similar problem or has a good idea on how to deal with
that.

It would mean that you would need a competent engineer at your site to
install, configure, and maintain the HylaFAX software. That engineer
would then have the freedom to read any of the source and would be
expected to publish any patches as a courtesy to the rest of us.  The
software includes licensing information, on how the software can be
used for commercial purposes but the code cannot be copied wholesale
and used commercially without providing source code and acknowledging
the copyrights of the original authors.

There is a tremendous amount of UNIX software published under similar
copyrights, so it is not unusual. The Linux operating system, the GNU
C++ compiler, the EMACS editor, X11 windowing system, Kerberos
authentication systems, PGP encryption code, etc. have all had similar
copyrights. And these systems really do *work* for people who need
robust and powerful tools.

Perhaps if you told us just how many computer systems of what
operating systems you need to work with, and what kind of clients for
your fax server you need, we could point you at the right tools, offer
ideas, or even suggest a consultant who could guide you through
configuration and help you get online?

If I lived in India, I'd consider doing it myself, but I think the
airfare would get expensive and I can't afford that much time right
now.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNLOskz/+ItycgIJRAQGBoQQAj3K1z2RVyw413D3gKFBz2D54H1B7haoO
eIXq6mX6mEw9a8UWuFxgRBCoKhDaICIDOhr2tk1eRQDmTFr6ohbfuj3hzWxGLKCs
lwK3JmJeLvBcWZHHe+4Po/cz5qB7Yl7rmM2zyDNcz8MalXzZRhPGxSC/C7cbPOPn
pOhyumFExm8=
=/ivI
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Comments: Authenticated sender is <spine001@gold.tc.umn.edu>
From: Julio C Spinelli <spine001@gold.tc.umn.edu>
To: Matthias.Apitz@SOFTCON.de
Date: Thu, 20 Nov 1997 21:47:09 +0000
Subject: Re: flexfax: What is Apointed Deadline ?
CC: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

Could somebody explain what service level of the sendpage command 
means?

Thanks in advance,

Julio

> Alessandro Martini wrote:
>     
>     Sometimes when i send a page by sendpage command, the Hylafax don't send
>     the page, the faxstat -s command say to me "Modem ..... running and idle",
>     and the Facsimile Agent say:
>      Your page job to 4223083 could not be completed befor the appointed deadline.
>     
>     What i do  ???
>     
> Check the hfaxd.conf file for the parameters PriorityMap:
> RetryTimeMap: and KillTimeMap: and compare it with the
> service level used in your sendpage(1) command.
> 
> 	matthias
> 
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: What is Apointed Deadline ?
To: spine001@gold.tc.umn.edu (Julio C Spinelli)
Date: Fri, 21 Nov 1997 07:45:10 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Julio C Spinelli wrote:
    
    Could somebody explain what service level of the sendpage command 
    means?
    
    Thanks in advance,

Just check the man page of sendpage(1) for the flag "-l" (and
if you want more information the RFC 1861 for SNPP).

	matthias
    
    > Alessandro Martini wrote:
    >     
    >     Sometimes when i send a page by sendpage command, the Hylafax don't send
    >     the page, the faxstat -s command say to me "Modem ..... running and idle",
    >     and the Facsimile Agent say:
    >      Your page job to 4223083 could not be completed befor the appointed deadline.
    >     
    >     What i do  ???
    >     
    > Check the hfaxd.conf file for the parameters PriorityMap:
    > RetryTimeMap: and KillTimeMap: and compare it with the
    > service level used in your sendpage(1) command.
    > 
    > 	matthias
    > 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 20 Nov 1997 20:18:38 -0500 (EST)
To: Nico Garcia <raoul@cirl.meei.harvard.edu>
From: Aidan Van Dyk <aidan@magma.ca>
Subject: Re: Re: flexfax: Fax != uucp
Cc: Hylafax Mailing list <flexfax@sgi.com>
Sender: owner-flexfax@celestial.com

On Thu, 20 Nov 1997 16:43:32 -0500 (EST), Nico Garcia <raoul@cirl.meei.harvard.edu> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> 
> On Fri, 21 Nov 1997, Jonathan Chen wrote:
> 
> > 
> > This is such a good idea! It will definitely solve the confusion
> > involved with installation, and the stubborness of some O/S's to allow
> > entries with the same uids.
> > 
> > A quick squizz thru' the source code reveals that it shouldn't really
> > be a problem (I'll try this in practise for a week on our system); in
> > fact "config.h" has the line:
> > 
> > 	#define  FAX_USER        "uucp"  /* account name of the ``fax user'' */
> > 
> > What do the others think?
> 
> This needs to operate on systems where uucp does and does not already
> exist. UUCP requires its own home directory: if the fax servers rely
> on the fax home directory to be distinct, this could cause serious
> grief.
> 
> I'd have to take a look at the chroot behavior of hfaxd and faxq to
> see if it could tolerate using the same home directory as UUCP. Has
> anyone double checked this?

Haven't tried it yet, but looking at the source, it uses
#define FAX_SPOOLDIR    "/var/spool/fax"        /* pathname to top of spooling area */

In config.h, and the they chdir to begin, then chroot(".").

In my quick search pwd->pw_dir is only used in client apps, in which case it is using the client's home directory.

I would think it should work, but I like to have a fax user, so I can do "ls ~fax/docq" types of things ;-)

Aidan.

--
Aidan Van Dyk                       aidan@magma.ca
-------------------------------------------------------------------
Magma Communications Ltd.           Home Page:  http://www.magma.ca
Unit 201, Antares Plaza             
52 Antares Drive                    Phone:      +1 613 228-3565
Nepean, ON, K2E 7Z1  CANADA         Fax:        +1 613 228-8313

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Leslie Mikesell <les@Mcs.Net>
Subject: Re: flexfax: Fax != uucp
To: jonc@pinnacle.co.nz (Jonathan Chen)
Date: Thu, 20 Nov 1997 09:52:55 -0600 (CST)
Cc: sam@campbellsci.co.uk, flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> > On my system though, I can't set the UID of "fax" to that of "uucp". 
> > 
> > Is this really important?
> 
> Yes, it is. What you've got to do is to twiddle /etc/passwd so that
> the uids are the same, and then also look at /tcb/files/auth/f/fax
> and alter the u_id# entry so that it matches uucp's (which should be 5
> on SCO OS 5.0.x).
> 
> Then run `/tcb/bin/authck -av' just to make sure everything's ok with
> the tcb subsystem.

Has anyone looked at the implications (or difficulty) of just changing
the name 'fax' as used internally by HylaFax to 'uucp' so this
wouldn't be an issue?  If you don't use uucp, it might be confusing
to see the name as the owner of files and processes, but if you
do, one or the other ends up being wrong any way you look at it.

  Les Mikesell
    les@mcs.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
X-Authentication-Warning: tui.pinnacle.co.nz: jonc owned process doing -bs
Date: Fri, 21 Nov 1997 09:07:39 +1300 (NZDT)
From: Jonathan Chen <jonc@pinnacle.co.nz>
To: Leslie Mikesell <les@Mcs.Net>
cc: Hylafax Mailing list <flexfax@sgi.com>
Subject: Re: flexfax: Fax != uucp
Sender: owner-flexfax@celestial.com

On Thu, 20 Nov 1997, Leslie Mikesell wrote:

> Has anyone looked at the implications (or difficulty) of just changing
> the name 'fax' as used internally by HylaFax to 'uucp' so this
> wouldn't be an issue?  If you don't use uucp, it might be confusing
> to see the name as the owner of files and processes, but if you
> do, one or the other ends up being wrong any way you look at it.

This is such a good idea! It will definitely solve the confusion
involved with installation, and the stubborness of some O/S's to allow
entries with the same uids.

A quick squizz thru' the source code reveals that it shouldn't really
be a problem (I'll try this in practise for a week on our system); in
fact "config.h" has the line:

	#define  FAX_USER        "uucp"  /* account name of the ``fax user'' */

What do the others think?
--
Jonathan Chen <jonc@pinnacle.co.nz>
----------------------------------------------------------------------
                                        "Nyuck, nyuck, nyuck!" - Curly

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 20 Nov 1997 16:43:32 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
Cc: Hylafax Mailing list <flexfax@sgi.com>
Subject: Re: flexfax: Fax != uucp
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Fri, 21 Nov 1997, Jonathan Chen wrote:

> On Thu, 20 Nov 1997, Leslie Mikesell wrote:
> 
> > Has anyone looked at the implications (or difficulty) of just changing
> > the name 'fax' as used internally by HylaFax to 'uucp' so this
> > wouldn't be an issue?  If you don't use uucp, it might be confusing
> > to see the name as the owner of files and processes, but if you
> > do, one or the other ends up being wrong any way you look at it.
> 
> This is such a good idea! It will definitely solve the confusion
> involved with installation, and the stubborness of some O/S's to allow
> entries with the same uids.
> 
> A quick squizz thru' the source code reveals that it shouldn't really
> be a problem (I'll try this in practise for a week on our system); in
> fact "config.h" has the line:
> 
> 	#define  FAX_USER        "uucp"  /* account name of the ``fax user'' */
> 
> What do the others think?

This needs to operate on systems where uucp does and does not already
exist. UUCP requires its own home directory: if the fax servers rely
on the fax home directory to be distinct, this could cause serious
grief.

I'd have to take a look at the chroot behavior of hfaxd and faxq to
see if it could tolerate using the same home directory as UUCP. Has
anyone double checked this?

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNHSuOz/+ItycgIJRAQEuUgP6Au5uiZBCrbrBey11uDqLtPond9ts71dK
3pEvSgvd/Nb0VjkTmdg3Nx2h894V5cWwE1nOD8oNzn3kiNTDXIVl3QMAFl2X+e6y
pZ7z6uUU7okJFhEiI4BDl+tAWasHbDCRnwBYOQlBMPloeeoc6Rgk16U+oHUFZ5vu
PJs1lGNsBBQ=
=x3iN
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Reply-To: "Chris Davis" <cjdavis@t-systems.com>
From: "Chris Davis" <cjdavis@t-systems.com>
To: <flexfax@sgi.com>
Cc: <uli@transcom.de>
Subject: flexfax: WHFC and problems with coverpages
Date: Thu, 20 Nov 1997 13:11:19 -0700
Sender: owner-flexfax@celestial.com

Hello,

I have managed to get HylaFAX up and running on a Linux box which will serve 3 to 5 Windows 95 clients. I am now trying to get cover pages worked into shape, and I have run into a snag. If I choose Use Faxcover, and select the standard faxcover.ps file, I get the following mail:

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 21 Nov 1997 08:22:56 +0100
From: Ulrich Eckhardt <uli@transcom.de>
Organization: Transcom
To: Chris Davis <cjdavis@t-systems.com>
CC: flexfax@sgi.com
Subject: Re: flexfax: WHFC and problems with coverpages
Sender: owner-flexfax@celestial.com

Chris Davis wrote:
> 
> Hello,
> 
> I have managed to get HylaFAX up and running on a Linux box which will serve 3 to 5 Windows 95 clients. I am now trying to get cover pages worked into shape, and I have run into a snag. If I choose Use Faxcover, and select the standard faxcover.ps file, I get the following mail:

Hi,

i don't know what mail you get, because i got only a part of your
e-mail,
but whfc has a bug in the coverpage generation. I'll fix it in the
next release (yes really i am working currently on it).

Uli 
-- 
Ulrich Eckhardt		                   mailto:uli@transcom.de
http://people.frankfurt.netsurf.de/uli
Truly great madness can not be achieved without significant
intelligence. (Henrik Tikkanen)


From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 20 Nov 1997 07:59:44 -0500
From: Aleksandar Simic <ASimic@compuserve.com>
Subject: Re: flexfax: 550: /tmp/doc5.ps: Not Owner
To: Tim Rice <tim@trr.metro.NET>
Cc: "flexfax@sgi.com" <flexfax@sgi.com>
Sender: owner-flexfax@celestial.com

Hello Tim, Everybody,

>> I'm having problems sending faxes. Error message is:
>> 
>>         550: /tmp/doc5.ps Not Owner
>> 

Tim Rice wrote:

>I don't run HP-UX, but I tracked down the same problem on SCO OS5
>and we fixed it in pl1.

>Check your port.h for CONFIG_MAXGID. If it is set to 60002, find
>out what went wrong with configure.
>THe relevent lines are.

>    # Select the maximum fax UID.  This value is stored
>    # in the GID of files so must be constrained by any
>    # system limitations.  The maximum value is used for
>   # anonymous logins.  The default value of 60002 comes
>    # from NFS/nobody usage on various systems.
>    #
>    test "$CONFIG_MAXGID" || {
>       case $TARGET in
>       *-sco*)         CONFIG_MAXGID=60000;;
>       *-hpux*)        CONFIG_MAXGID=60000;;
>       *)              CONFIG_MAXGID=60002;;   # default
>       esac
>    }
> 

Thanks for your suggestion. I did check and it seems that HP-UX should be
OK
with MAXGID of 60000. I aslo checked config.log and couldn't find anything
on 
GID, nor chown nor chgrp nor chmod. 

Also my fax user has UID:GID 3:5 exactly the same as uucp user. 
At this time I'm gettig pretty desperate ;-(
So if anybody else would have any other idea, I'd really appreciate it.

Thanks,
Aleksandar Simic (asimic@4compsys.com, asimic@compuserve.com)
 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: 21 Nov 1997 07:43:51 -0000
From: Felix Morley Finch <felix@crowfix.com>
To: flexfax@sgi.com
Subject: flexfax: Same computer, new phone line, no dial tone
Sender: owner-flexfax@celestial.com

I moved, got another phone line, and using the same computer, same
software, it no longer dials out.  Complains about NO DIAL TONE.  I'm
pretty sure it's a modem problem because cu with a manual atdt1234567
has the same problem.  A phone works fine whether routed thru the
modem card or not, and the NO DIAL TONE error shows up whether the
phone is routed thru the modem card or not connected at all.

I can hear it go off hook very briefly, then NO DIAL TONE.

It's a several year old USR COURIER.

I have an ISDN line for the net connection, so there should be no
problem there.

Could the phone line quality somehow be causing this?  Or does anybody
have any other clues?

-- 
            ... _._. ._ ._. . _._. ._. ___ .__ ._. . .__. ._ .. ._.
      Felix Finch, scarecrow repairer, rocket surgeon / felix@crowfix.com
            PGP = 91 B3 94 7C E9 E8 76 2D   E1 63 51 AA A0 48 89 2F
I've found a solution to Fermat's Last Theorem but I see I've run out of room o

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 21 Nov 1997 12:23:30 +0100 (CET)
From: Matjaz Godec <gody@elgo.si>
To: Felix Morley Finch <felix@crowfix.com>
cc: flexfax@sgi.com
Subject: Re: flexfax: Same computer, new phone line, no dial tone
Sender: owner-flexfax@celestial.com

On 21 Nov 1997, Felix Morley Finch wrote:

Try ATX3DT...

	lp
		gody
___________________________________________________________________________
Matjaz Godec            |     "ELGO" Maribor d.o.o.   |     Electronics
matjaz.godec@elgo.si    |     Ul. Josipa Priola 35    | Telecommunications
tel/fax:+ 386 62 214460 |  si 2ooo Maribor, Slovenija |    CALDERA-LINUX
http://www.elgo.si/	|   Slovenian network "SloN"  | Internet consultant

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Same computer, new phone line, no dial tone
To: felix@crowfix.com (Felix Morley Finch)
Date: Fri, 21 Nov 1997 14:04:26 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Felix Morley Finch wrote:
    
    I moved, got another phone line, and using the same computer, same
    software, it no longer dials out.  Complains about NO DIAL TONE.  I'm
    pretty sure it's a modem problem because cu with a manual atdt1234567
    has the same problem.  A phone works fine whether routed thru the
    modem card or not, and the NO DIAL TONE error shows up whether the
    phone is routed thru the modem card or not connected at all.
    
    I can hear it go off hook very briefly, then NO DIAL TONE.
    
    It's a several year old USR COURIER.
    
    I have an ISDN line for the net connection, so there should be no
    problem there.

ISDN doesn't have such things like "tones". It has a D-channel
for call-setup with messages. It's the task of a device like
a ISDN-PBX or a simple a/b adapter to chat D-channel messages
with the telco and provide  the appropriate analog voltage
signals (like dial tone or ring voltage) to the connected analog
phone or modem.

If you have changed this device (because you have now ISDN) the dial
tone voltage or frequency may also have changed and the modem
can't detect the dial tone any more. Most modems support an
ATX? command (e.g. ATX3) to ignore this problem and to go on
with dialing anyway. Check the manual of your modem and add
this value to ModemDialCmd: in the modem config file.

If you don't have such a device like ISDN-PBX or a/b adapter
and only a plain so called So (read S-null) bus you can't
connect your modem and must switch to an ISDN-modem/TA hybrid
(e.g. to a ZyXEL Elite 2864I or to ISDN COURIER modem card).

	matthias
    
    Could the phone line quality somehow be causing this?  Or does anybody
    have any other clues?
    
    -- 
                ... _._. ._ ._. . _._. ._. ___ .__ ._. . .__. ._ .. ._.
          Felix Finch, scarecrow repairer, rocket surgeon / felix@crowfix.com
                PGP = 91 B3 94 7C E9 E8 76 2D   E1 63 51 AA A0 48 89 2F
    I've found a solution to Fermat's Last Theorem but I see I've run out of room o

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 21 Nov 1997 08:59:16 -0600 (CST)
From: Bill Suetholz <bill@centonline.com>
To: Felix Morley Finch <felix@crowfix.com>
cc: flexfax@sgi.com
Subject: Re: flexfax: Same computer, new phone line, no dial tone
Sender: owner-flexfax@celestial.com

I've seen similar behavior in other modems when you have the phone line
plugged into the phone port instead of the line port on the modem.

Just to point out a simple fix...

Bill Suetholz


>On 21 Nov 1997, Felix Morley Finch wrote:
>
> I moved, got another phone line, and using the same computer, same
> software, it no longer dials out.  Complains about NO DIAL TONE.  I'm
> pretty sure it's a modem problem because cu with a manual atdt1234567
> has the same problem.  A phone works fine whether routed thru the
> modem card or not, and the NO DIAL TONE error shows up whether the
> phone is routed thru the modem card or not connected at all.
> 
> I can hear it go off hook very briefly, then NO DIAL TONE.
> 
> It's a several year old USR COURIER.
> 
> I have an ISDN line for the net connection, so there should be no
> problem there.
> 
> Could the phone line quality somehow be causing this?  Or does anybody
> have any other clues?
> 
> -- 
>             ... _._. ._ ._. . _._. ._. ___ .__ ._. . .__. ._ .. ._.
>       Felix Finch, scarecrow repairer, rocket surgeon / felix@crowfix.com
>             PGP = 91 B3 94 7C E9 E8 76 2D   E1 63 51 AA A0 48 89 2F
> I've found a solution to Fermat's Last Theorem but I see I've run out of room o
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: 21 Nov 1997 15:58:44 -0000
From: Felix Morley Finch <felix@crowfix.com>
To: flexfax@sgi.com
Subject: Re: flexfax: Same computer, new phone line, no dial tone -- SOLVED
Sender: owner-flexfax@celestial.com

I was about to send off a condescending answer to this that, no, it
can't be, because a regular phone works fine when routed thru the
computer, so no, it can't be having the plugs swapped.  Then I
thought, well, always check, never assume, and BINGO!  It works.

>>In article <Pine.SOL.3.96.971121090649.7532B-100000@copland.udel.edu>, Mike Porter <mike@UDel.Edu> writes:

> I've done trhis before.  When I did it, it was because I plugged
> the phone line into the wrong jack on the modem.  Ie:  The wall
> line plugged into the telephone jack on the modem instead of
> the 'wall line' jack.

I had written:

>> A phone works fine whether routed thru the
>> modem card or not, and the NO DIAL TONE error shows up whether the
>> phone is routed thru the modem card or not connected at all.

So now I guess I'll just have have another helping of humble pie.
Thanks Mike!

-- 
            ... _._. ._ ._. . _._. ._. ___ .__ ._. . .__. ._ .. ._.
      Felix Finch, scarecrow repairer, rocket surgeon / felix@crowfix.com
            PGP = 91 B3 94 7C E9 E8 76 2D   E1 63 51 AA A0 48 89 2F
I've found a solution to Fermat's Last Theorem but I see I've run out of room o

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 21 Nov 1997 08:25:34 +0000
To: HylaFAX List <flexfax@sgi.com>
From: Phil Watkinson <pkw@elgro.co.uk>
Subject: Re: flexfax: Mail Gateway problems
Cc: Steve <steve@sjs.com>
Sender: owner-flexfax@celestial.com

At 15:02 20/11/1997 -0500, Steve Stuczynski wrote:
>	I have hylafax-v4.0pl1 compiled and installed in a Slackware V3.4
>Linux V2.0.30 system. Using command-line "sendfax" command, I can send
>faxes just fine.
>	When I try to send a fax by sending a fax via e-mail this way:
>
>To      : user@host.FAX
>Cc      : 
>Attchmnt: fax.001
>Subject : lc 16171231234 steve@sjs.com
>
>
>	The logfile shows that it's trying to dial 8737 . Nomatter what I
>put in the subject, it tries to dial 8737.....

Steve, 

HylaFAX is expecting the 'To' address to be in the form 'user@123456.FAX'
so it is trying to dial 'host' - ie 8737. Geddit?

Have a go with 'To : steve@16171231234.FAX' and you can put anything you
like in the Subject line.

I note you are trying to fax an attachment called 'fax.001'. If you need
this capability I would recommend careful reading of the faxmail man
page; otherwise stick to putting your messages in the email body.

Cheers,
Phil Watkinson,
Boston, UK.

From steve@sjs.com  Wed Feb 17 13:55:48 1999
Date: Fri, 21 Nov 1997 10:50:29 -0500 (EST)
From: Steve <steve@sjs.com>
To: Phil Watkinson <pkw@elgro.co.uk>
cc: HylaFAX List <flexfax@sgi.com>
Subject: Re: flexfax: Mail Gateway problems

Hello Phil,
	I did as you said, I send the message to user@16174555288.FAX on
the Fax server system, but it still doesn't fax..
	Response from faxstat:

# faxstat
HylaFAX scheduler on www.metwest.com: Running

	The docq directory has two entries from the two e-mail tests I
tried:

www:/usr/spool/fax/docq# ls -la
total 11
drwx------   2 uucp     uucp         1024 Nov 21 10:42 ./
drwxr-xr-x  17 uucp     uucp         1024 Nov 18 14:18 ../
-rw-rw----   1 uucp     60002        3743 Nov 21 10:25 doc1.ps.14
-rw-rw----   1 uucp     60002        3746 Nov 21 10:42 doc2.ps.15
-rw-------   1 uucp     60002           2 Nov 21 10:42 seqf


and nothing is in the log directory:

www:/usr/spool/fax/log# ls -la
total 3
drwxr-xr-x   2 uucp     uucp         1024 Nov 21 10:24 ./
drwxr-xr-x  17 uucp     uucp         1024 Nov 18 14:18 ../
-rw-------   1 uucp     wheel           2 Nov 21 10:24 seqf


This is what I added to my sendmail.cf file:

######################################   
###   Ruleset 0 -- Parse Address   ###
######################################
 
S0

# # forward FAX messages to HylaFAX software
R$+<@$+.FAX>    $#fax $@ $2 $: $1                       user@host.FAX


	And

######################################################################
######################################################################
#####
#####                   MAILER DEFINITIONS
#####
######################################################################
######################################################################

#
Mfax,   P=/usr/local/lib/fax/mailfax, F=DFMShu, M=100000,
        A=mailfax $u $h $f
#


I'm lost!!!

Thank you,
	Steve Stuczynski

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 21 Nov 1997 17:02:06 +0000
To: Steve <steve@sjs.com>
From: Phil Watkinson <pkw@elgro.co.uk>
Subject: Re: flexfax: Mail Gateway problems
Cc: HylaFAX List <flexfax@sgi.com>
Sender: owner-flexfax@celestial.com

At 10:50 21/11/1997 -0500, you wrote:
>Hello Phil,
>	I did as you said, I send the message to user@16174555288.FAX on
>the Fax server system, but it still doesn't fax..
>[snip]
>This is what I added to my sendmail.cf file:
>
>######################################   
>###   Ruleset 0 -- Parse Address   ###
>######################################
> 
>S0
>
># # forward FAX messages to HylaFAX software
>R$+<@$+.FAX>    $#fax $@ $2 $: $1                       user@host.FAX
>
>[snip-ola]

Well my sendmail.cf says different :

# 
#  # resolve fake top level domains by forwarding to other hosts
#  R$+< @ $+ .FAX. >		$#fax $@ $2 $: $1	user@host.FAX
#

But you have to be ever so careful with this file because spaces and tabs do
make a difference. Much better to amend the m4 macro file, which is probably
called metwest.mc on your machine and then regenerate the sendmail.cf file
from it.

If you need more help with this, there are full-ish instructions at :
		http://www.elgro.demon.co.uk/hylafax/faxgateway.html

BTW, and it's a silly question, but have you remembered the 
/usr/local/lib/fax/mailfax script?

Good Luck,
Phil Watkinson,
Boston, UK.

From steve@sjs.com  Wed Feb 17 13:55:48 1999
Date: Fri, 21 Nov 1997 13:46:37 -0500 (EST)
From: Steve <steve@sjs.com>
To: Phil Watkinson <pkw@elgro.co.uk>
Subject: Re: flexfax: Mail Gateway problems

Hi Phil,
	Now I'm really lost!!! I followed all the directions
mailfax.sh-sendmail file and started the faxq daemon. When I submit a fax
via e-mail or using the sendfax program, files get generated in the
/usr/spool/fax/docq directory but not logfiles appear in the
/usr/spool/fax/log directory.

a faxstat shows:

HylaFAX scheduler on www.metwest.com: Running

But no faxes get sent.
	Atleast the other day sendfax would actually send a fax out.

Anyother ideas?

Thanks,
	Steve

From steve@sjs.com  Wed Feb 17 13:55:48 1999
Date: Fri, 21 Nov 1997 16:30:06 -0500 (EST)
From: Steve <steve@sjs.com>
To: Phil Watkinson <pkw@elgro.co.uk>
Subject: Re: flexfax: Mail Gateway problems

Hi Phil,
	O.k., it looks to me like your code in the sendmail.cf is
commented out, am I right? And are those spaces in the < >? I added the
spaces inbetween the chars in the <> and it didn't help. I also watch my
syslog file when I restart the sendmail daemon looking for sendmail.cf
errors and I don't see any.
	I hate playing around with the sendmail.cf file, it's such a
pain...

Thank you,
	Steve


On Fri, 21 Nov 1997, Phil Watkinson wrote:

> At 10:50 21/11/1997 -0500, you wrote:
> >Hello Phil,
> >	I did as you said, I send the message to user@16174555288.FAX on
> >the Fax server system, but it still doesn't fax..
> >[snip]
> >This is what I added to my sendmail.cf file:
> >
> >######################################   
> >###   Ruleset 0 -- Parse Address   ###
> >######################################
> > 
> >S0
> >
> ># # forward FAX messages to HylaFAX software
> >R$+<@$+.FAX>    $#fax $@ $2 $: $1                       user@host.FAX
> >
> >[snip-ola]
> 
> Well my sendmail.cf says different :
> 
> # 
> #  # resolve fake top level domains by forwarding to other hosts
> #  R$+< @ $+ .FAX. >		$#fax $@ $2 $: $1	user@host.FAX
> #
> 
> But you have to be ever so careful with this file because spaces and tabs do
> make a difference. Much better to amend the m4 macro file, which is probably
> called metwest.mc on your machine and then regenerate the sendmail.cf file
> from it.
> 
> If you need more help with this, there are full-ish instructions at :
> 		http://www.elgro.demon.co.uk/hylafax/faxgateway.html
> 
> BTW, and it's a silly question, but have you remembered the 
> /usr/local/lib/fax/mailfax script?
> 
> Good Luck,
> Phil Watkinson,
> Boston, UK.
> 

From steve@sjs.com  Wed Feb 17 13:55:48 1999
Date: Fri, 21 Nov 1997 16:55:37 -0500 (EST)
From: Steve <steve@sjs.com>
To: Phil Watkinson <pkw@elgro.co.uk>
Subject: Re: flexfax: Mail Gateway problems

Hello Phil,

I just cut exactly what you sent me for your sendmail.cf into my
sendmail.cf:

######################################
###   Ruleset 0 -- Parse Address   ###
######################################

S0

#
#  # resolve fake top level domains by forwarding to other hosts
#  R$+< @ $+ .FAX. >    $#fax $@ $2 $: $1       user@host.FAX
#

I restarted sendmail, looked in syslog and messages logfiles, no warnings
or errors and it didn't send a fax. It placed a doc9.ps into
/usr/spool/fax/docq and that's it. Nothing in /usr/spool/fax/log

What else can I try? Could you send me your whole sendmail.cf as an
attachment?

Thank you,
	Steve

P.S. Once I finally get it working how do other machines on my Lan or off
my Lan send Faxmail? Do I create an alias account that points to
user@host.FAX?

From steve@sjs.com  Wed Feb 17 13:55:48 1999
Date: Fri, 21 Nov 1997 20:10:29 -0500 (EST)
From: Steve <steve@sjs.com>
To: Phil Watkinson <pkw@elgro.co.uk>
cc: HylaFAX List <flexfax@sgi.com>
Subject: Re: flexfax: Mail Gateway problems

Hello Phil,
	I got it working, thank you. The documentation that came with the
hylafax-v4.0pl1 for mailfax.sh-sendmail is incorrect!!!
	I found an old posting of yours and with a combination of your
posting and mailfax.sh-sendmail, I believe I got it working. I'll know
better tomorrow when I go to the site I faxed and getthe faxes.
	How can other machines on my Lan or off my Lan use the FaxServer?
Something like user@phone_number.fax@my_domain.com ?


	Here is what I added to my sendmail.cf file:

######################################   
###   Ruleset 0 -- Parse Address   ###
######################################
 
S0

# 
#  # resolve fake top level domains by forwarding to other hosts   
R$+< @ $+ .FAX >        $#fax $@ $2 $: $1       user@host.FAX
# 


and:

######################################################################
######################################################################
#####
#####                   MAILER DEFINITIONS
#####
######################################################################
######################################################################
  
#  
Mfax,   P=/usr/local/lib/fax/mailfax, F=DFMhu, S=14, R=24,   
        M=100000, T=X-Phone/X-FAX/X-Unix,
        A=mailfax $u $h $f
#

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 24 Nov 1997 09:45:08 +0000
To: HylaFAX List <flexfax@sgi.com>
From: Phil Watkinson <pkw@elgro.co.uk>
Subject: Re: flexfax: Mail Gateway problems
Cc: Steve <steve@sjs.com>
Sender: owner-flexfax@celestial.com

At 16:30 on Friday, Steve <steve@sjs.com> wrote:
>Hi Phil,

Hi, Steve. You've been busy.

>	O.k., it looks to me like your code in the sendmail.cf is
>commented out, am I right? 

Ooops, only because I pasted it from my mailfax.sh-sendmail script.
The point I was trying to make was that it's easy to make mistakes
if you manually edit the sendmail.cf file...

>	I hate playing around with the sendmail.cf file, it's such a
>pain...

Quite right. You must get used to using the m4 macro file instead.


And at 16:55 on Friday, you said:
>What else can I try? Could you send me your whole sendmail.cf as an
>attachment?
>

Certainly not. But I will send you directly my mc file to show you
that its 10 lines will generate 1000 lines of sendmail.cf - and you
will see the MAILER(fax) line which accurately does all the changes
you have been struggling with.

>P.S. Once I finally get it working how do other machines on my Lan or
>off my Lan send Faxmail? Do I create an alias account that points to
>user@host.FAX?
>

Well, its simpler than you think. Any client that can send email to the
server can use the email to fax gateway. All that is necessary is the
correct format of the 'To' address, exactly as you have been testing.
You do not need an sendmail alias account, altho' they can be convenient.
There is an example in the faxmail man page.


And At 20:10 Friday, you added:
>Hello Phil,
>	I got it working, thank you.

Phew! Alls well that's ends well.

> The documentation that came with the
>hylafax-v4.0pl1 for mailfax.sh-sendmail is incorrect!!!

Unfortunately so; I have submitted a patch to Matthias, and the html
documentation has been revised as well for v4.0pl2.

>	I found an old posting of yours and with a combination of your
>posting and mailfax.sh-sendmail, I believe I got it working. I'll know
>better tomorrow when I go to the site I faxed and getthe faxes.

I'll keep my fingers crossed.

Cheers,
Phil Watkinson,
Boston, UK.

From bill@centonline.com  Wed Feb 17 13:55:48 1999
Date: Mon, 24 Nov 1997 12:16:45 -0600 (CST)
From: Bill Suetholz <bill@centonline.com>
To: Phil Watkinson <pkw@elgro.co.uk>
Subject: Re: flexfax: Mail Gateway problems

Maybe you can post your m4 source to the list?

Bill Suetholz

>On Mon, 24 Nov 1997, Phil Watkinson wrote:
>
> At 16:30 on Friday, Steve <steve@sjs.com> wrote:
> >Hi Phil,
> 
> Hi, Steve. You've been busy.
> 
> >	O.k., it looks to me like your code in the sendmail.cf is
> >commented out, am I right? 
> 
> Ooops, only because I pasted it from my mailfax.sh-sendmail script.
> The point I was trying to make was that it's easy to make mistakes
> if you manually edit the sendmail.cf file...
> 
> >	I hate playing around with the sendmail.cf file, it's such a
> >pain...
> 
> Quite right. You must get used to using the m4 macro file instead.
> 
> 
> And at 16:55 on Friday, you said:
> >What else can I try? Could you send me your whole sendmail.cf as an
> >attachment?
> >
> 
> Certainly not. But I will send you directly my mc file to show you
> that its 10 lines will generate 1000 lines of sendmail.cf - and you
> will see the MAILER(fax) line which accurately does all the changes
> you have been struggling with.
> 
> >P.S. Once I finally get it working how do other machines on my Lan or
> >off my Lan send Faxmail? Do I create an alias account that points to
> >user@host.FAX?
> >
> 
> Well, its simpler than you think. Any client that can send email to the
> server can use the email to fax gateway. All that is necessary is the
> correct format of the 'To' address, exactly as you have been testing.
> You do not need an sendmail alias account, altho' they can be convenient.
> There is an example in the faxmail man page.
> 
> 
> And At 20:10 Friday, you added:
> >Hello Phil,
> >	I got it working, thank you.
> 
> Phew! Alls well that's ends well.
> 
> > The documentation that came with the
> >hylafax-v4.0pl1 for mailfax.sh-sendmail is incorrect!!!
> 
> Unfortunately so; I have submitted a patch to Matthias, and the html
> documentation has been revised as well for v4.0pl2.
> 
> >	I found an old posting of yours and with a combination of your
> >posting and mailfax.sh-sendmail, I believe I got it working. I'll know
> >better tomorrow when I go to the site I faxed and getthe faxes.
> 
> I'll keep my fingers crossed.
> 
> Cheers,
> Phil Watkinson,
> Boston, UK.
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 24 Nov 1997 09:44:46 +0100 (MET)
From: Carsten Hoeger <choeger@suse.de>
To: Jonathan Chen <jonc@pinnacle.co.nz>
cc: Leslie Mikesell <les@Mcs.Net>, Hylafax Mailing list <flexfax@sgi.com>
Subject: Re: flexfax: Fax != uucp
Sender: owner-flexfax@celestial.com

On Fri, 21 Nov 1997, Jonathan Chen wrote:

> On Thu, 20 Nov 1997, Leslie Mikesell wrote:
> 
> > Has anyone looked at the implications (or difficulty) of just changing
> > the name 'fax' as used internally by HylaFax to 'uucp' so this
> > wouldn't be an issue?  If you don't use uucp, it might be confusing
> > to see the name as the owner of files and processes, but if you
> > do, one or the other ends up being wrong any way you look at it.
> 
> This is such a good idea! It will definitely solve the confusion
> involved with installation, and the stubborness of some O/S's to allow
> entries with the same uids.
> 
> A quick squizz thru' the source code reveals that it shouldn't really
> be a problem (I'll try this in practise for a week on our system); in
> fact "config.h" has the line:
> 
> 	#define  FAX_USER        "uucp"  /* account name of the ``fax user'' */
> 
> What do the others think?

I did not change the defined FAX_USER variable, but I changed the
setup-scripts faxsetup and faxaddmodem and I gave fax another uid than
uucp and it works...

mfG,
	Carsten Hoeger
------
Carsten Hoeger  - S.u.S.E. GmbH -  Gebhardtstr. 2  -  90762 Fuerth  -  Germany
fax +49-911-3206727      web http://www.suse.de        mailbox +49-911-3247122
"Ein Buch ist eine Wolke, auf der man keinen Husten bekommt." (Max Goldt)

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Leslie Mikesell <les@mcs.net>
Subject: Re: flexfax: Fax != uucp
To: choeger@suse.de (Carsten Hoeger)
Date: Mon, 24 Nov 1997 12:25:26 -0600 (CST)
Cc: jonc@pinnacle.co.nz, les@mcs.net, flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> On Fri, 21 Nov 1997, Jonathan Chen wrote:
> 
> > > Has anyone looked at the implications (or difficulty) of just changing
> > > the name 'fax' as used internally by HylaFax to 'uucp' so this
> > > wouldn't be an issue?

> > 	#define  FAX_USER        "uucp"  /* account name of the ``fax user'' */
> > What do the others think?
> 
> I did not change the defined FAX_USER variable, but I changed the
> setup-scripts faxsetup and faxaddmodem and I gave fax another uid than
> uucp and it works...

The problem here is that you can't share usage of the modems with any
other programs (uucp, cu, kermit, etc.) that depend on the modem
device being writable by uucp's uid.  If you use any such programs
the uid has to match or the system permissions won't work.  The
text of the name is only cosmetic.

  Les Mikesell
  les@mcs.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Tue, 25 Nov 1997 10:11:24 +0100 (MET)
From: Carsten Hoeger <choeger@suse.de>
To: Leslie Mikesell <les@mcs.net>
cc: jonc@pinnacle.co.nz, flexfax@sgi.com
Subject: Re: flexfax: Fax != uucp
Sender: owner-flexfax@celestial.com

On Mon, 24 Nov 1997, Leslie Mikesell wrote:

> > On Fri, 21 Nov 1997, Jonathan Chen wrote:
> > 
> > > > Has anyone looked at the implications (or difficulty) of just changing
> > > > the name 'fax' as used internally by HylaFax to 'uucp' so this
> > > > wouldn't be an issue?
> 
> > > 	#define  FAX_USER        "uucp"  /* account name of the ``fax user'' */
> > > What do the others think?
> > 
> > I did not change the defined FAX_USER variable, but I changed the
> > setup-scripts faxsetup and faxaddmodem and I gave fax another uid than
> > uucp and it works...
> 
> The problem here is that you can't share usage of the modems with any
> other programs (uucp, cu, kermit, etc.) that depend on the modem

I can, because all the fax-stuff runs with uid uucp. The fax user is in
the same group (uucp) as uucp is (what else). And /var/lock is set rwx for
the group uucp.

> device being writable by uucp's uid.  If you use any such programs
> the uid has to match or the system permissions won't work.  The
> text of the name is only cosmetic.



mfG,
	Carsten Hoeger
------
Carsten Hoeger  - S.u.S.E. GmbH -  Gebhardtstr. 2  -  90762 Fuerth  -  Germany
fax +49-911-3206727      web http://www.suse.de        mailbox +49-911-3247122
"Ein Buch ist eine Wolke, auf der man keinen Husten bekommt." (Max Goldt)

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Tue, 25 Nov 1997 11:35:30 -0500 (EST)
From: Nico Garcia <raoul@cirl.meei.harvard.edu>
Cc: Nico Garcia <flexfax@sgi.com>
Subject: Re: flexfax: Fax != uucp
Sender: owner-flexfax@celestial.com

-----BEGIN PGP SIGNED MESSAGE-----

On Tue, 25 Nov 1997, Carsten Hoeger wrote:

> > > I did not change the defined FAX_USER variable, but I changed the
> > > setup-scripts faxsetup and faxaddmodem and I gave fax another uid than
> > > uucp and it works...
> > 
> > The problem here is that you can't share usage of the modems with any
> > other programs (uucp, cu, kermit, etc.) that depend on the modem
> 
> I can, because all the fax-stuff runs with uid uucp. The fax user is in
> the same group (uucp) as uucp is (what else). And /var/lock is set rwx for
> the group uucp.

And the device needs to be set "rw" for group uucp as well. And any
new modems or devices have to be set up at boot time. And *all*
fax/modem software has to use the same permissions. And....

Like changing from English screws to metric, it involves a lot of
changes that have to propagate correctly elsewhere. I'm not convinced
it's worth the grief, particularly if HylaFAX is to remain compatible
with the broadest range of other software.

			Nico Garcia
			Engineer, CIRL 
			Mass. Eye and Ear Infirmary
			raoul@cirl.meei.harvard.edu

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBNHr+SD/+ItycgIJRAQFFhQP/VJQGbVzUNOzFzzWKrph/yYbeQ17YX+L5
0XkVtxHuIAvzCO4XvhcO4H/4Or3AaJrSHbeE0bycbzNUWXCBe2OQJrV4b3Y5+pbT
BhBFpSAA/GAYzxW/Z3ExezR1v9jJRIMnDOzAQsisoVwnp11occERdZwjn1AYB6rR
1VxpgyB8ErE=
=18aK
-----END PGP SIGNATURE-----

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Leslie Mikesell <les@mcs.net>
Subject: Re: flexfax: Fax != uucp
To: choeger@suse.de (Carsten Hoeger)
Date: Tue, 25 Nov 1997 12:14:36 -0600 (CST)
Cc: les@mcs.net, jonc@pinnacle.co.nz, flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> > > I did not change the defined FAX_USER variable, but I changed the
> > > setup-scripts faxsetup and faxaddmodem and I gave fax another uid than
> > > uucp and it works...
> > 
> > The problem here is that you can't share usage of the modems with any
> > other programs (uucp, cu, kermit, etc.) that depend on the modem
> 
> I can, because all the fax-stuff runs with uid uucp. The fax user is in
> the same group (uucp) as uucp is (what else). And /var/lock is set rwx for
> the group uucp.

The device itself is the main problem.  Unless you have seriously
malicious users you can probably set /var/lock to mode 777.  But
you probably want to limit access to the modems and cu/kermit, etc.
know how to setuid uucp but I don't think they know much about groups.
Also, on at least some systems if you ever run the system uugetty
or ttymon it will change the owner to uucp, mode to rw only by owner.
If you can keep group rw enabled on the device and leave it owned
by the uucp uid everything should work.

  Les Mikesell
   les@mcs.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 24 Nov 1997 08:48:52 +0100
From: Ulrich Eckhardt <uli@transcom.de>
Organization: Transcom
To: HylaFax Mailing liste <flexfax@sgi.com>
Subject: Re: flexfax: WHFC and problems with coverpages
Sender: owner-flexfax@celestial.com

Chris Davis wrote:
> 
> Well, crud. As in, this microsoft outlook email client is crud. Sorry about
> that.
> 
> The rest of the mail that was chopped off was supposed to say that I had
> seen on the WHFC web page that cover page support was broken so I
> was not surprised when it didn't work. The questions I have are:
> 
> - Is there any sort of workaround? The web page said it won't work on
>   'standard' cover pages - is there a 'non-standard' page that will work?
>   We will only be sending faxes from WHFC on the win95 machines, if
>   this helps.
> 
> - When should I expect a new version of WHFC? I don't mean to push - I
>   really like the software!

Hi,

add the following fragment at the beginning of your cover page :

/IS {
   dup where {
      pop
      load dup nullstring ne {S} {pop} ifelse
   } {pop} ifelse
} bind def
/M { moveto } bind def
/S { show } bind def

and also don't use the comment1() to commentx() lines, use only the
comment() macro.

But i would recomment to wait untill the next release is out. But this
can take some time, because i have to write something that's called
a port monitor for the problems with Windows 95 and problems when
sending
large faxes with the fax printer feature. 

Uli
-- 
Ulrich Eckhardt		                   mailto:uli@transcom.de
http://people.frankfurt.netsurf.de/uli
Truly great madness can not be achieved without significant
intelligence. (Henrik Tikkanen)

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: "McNitt, Jennifer" <Jennifer.McNitt@westgroup.com>
To: "'flexfax@sgi.com'" <flexfax@sgi.com>
Subject: FW: flexfax: Turn off cover page
Date: Fri, 21 Nov 1997 09:44:02 -0600
Sender: owner-flexfax@celestial.com



> -----Original Message-----
> From:	McNitt, Jennifer 
> Sent:	Friday, November 21, 1997 9:43 AM
> To:	'Julio C Spinelli'
> Cc:	'felxfax@sgi.com'
> Subject:	RE: flexfax: Turn off cover page
> 
> I consulted OM's support group about your reply and they do not have a
> call to sendpage in their source but they do have a call to a program
> called sendfax.  My support person was wondering if sendfax calls
> sendpage.  Here is a snipet of the code that they sent me:
> 
> proc accept_fax_cgi {} {
> 
>     set current_dir [pwd]
>     set fax_dir /opt/tms/tms-fax-s
>     set env(PATH) \
>         $env(PATH):$fax_dir/hylafax/local/bin:$fax_dir/hylafax/lib/fax
> 
>     if [catch {exec $fax_dir/hylafax/local/bin/sendfax \
>                -h localhost \
>                -d $fields(recipient)@$fields(phonenum) \
>                -f $fields(sender) \
>                -r \"$fields(regarding)\" \
>                $fileName} err] {
>         puts       "Failed.
>             Encountered error $err while trying to send fax to
>             $fields(recipient). Phone number $fields(phonenum).
>             Message is stored in file $fileName"
>         syslog_log "Encountered error $err while trying to send fax to
>             $fields(recipient). Phone number $fields(phonenum).
>             Message is stored in file $fileName"
>     } else {
>         puts "Success.
>           Your request has been queued for transmission.
>           The transmission results are: $err"
>         syslog_log "Submitted fax for transmission to
>             $fields(recipient). Phone number $fields(phonenum).
>             The transmission results are: $err"
> 	...
> 	}
> 
> Do you know if I will be able to turn of the cover page with the call
> to sendfax hardcoded?
> 
> Thanks,
> 
> Jen
> 
> -----Original Message-----
> From:	Julio C Spinelli [SMTP:spine001@gold.tc.umn.edu]
> Sent:	Thursday, November 20, 1997 2:43 AM
> To:	McNitt, Jennifer
> Subject:	Re: flexfax: Turn off cover page
> 
> Just do sendpage -n -dxxxxxxx, make sure the -n is before the -d, it 
> won't send your cover page.
> 
> Julio
> > I'm using Hylafax with Open Market's Transact product.  I would like
> to
> > modify the setup so that it does not create a cover sheet for each
> fax.
> > Is there a config file change that I can make do to this?
> > 
> > Thanks,
> > 
> > Jen
> > 
> > ____
> > Jennifer McNitt
> > West Group
> > 612-687-5273
> > jennifer.mcnitt@westgroup.com
> > 
> > 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
To: flexfax@sgi.com
Subject: flexfax: hylafax 4.0pl1 can't send fax to certain fax machines
X-Face: '%)H7X3>"mU[aaPz3aZ`^e{*DV1&W;+P-Wx*Yk)Ah[e(-AO&EB\D.gq2y{kH73Q29}_A4H,eykdf!9dj[m\gwFgZOTt?=7gI+RBN,I({A32-MyZrmtrD%8*L/boGS0,+Lj$H|@
X-Url: http://www.francenet.fr/
Date: Fri, 21 Nov 1997 17:08:52 +0100
From: Gildas Perrot <perrot@francenet.fr>
Sender: owner-flexfax@celestial.com

Hi everybody,

I am using hylafax 4.0pl1 on BSD/OS 2.1 to send faxes with USR Robotics 
Courier V34bis. Usually it works well but with some fax machines, I have 
problems. They receive the call of my hylafax modem but the fax is not send. 
Below are the session logs.
Any idea about that problem ?

Thanks a lot in advance for your help.		Gildas.

Your facsimile job to 3610,8120767246029,0231884129# failed because:
    No answer (T.30 T1 timeout); too many attempts to send

    ---- Transcript of session follows ----

Nov 19 16:49:58.56: [20798]: SESSION BEGIN 00006313 
331361081207672460290231884129
Nov 19 16:49:58.56: [20798]: SEND FAX: JOB 4061 DEST 
3610,8120767246029,0231884129# COMMID 00006313
Nov 19 16:49:58.56: [20798]: DELAY 2600 ms
Nov 19 16:50:01.17: [20798]: <-- [17:ATE0V1Q0X4S0=0H0\r]
Nov 19 16:50:01.33: [20798]: --> [16:ATE0V1Q0X4S0=0H0]
Nov 19 16:50:01.33: [20798]: --> [2:OK]
Nov 19 16:50:01.33: [20798]: <-- [32:ATS8=2S7=60&H2&I2&R1S13=1&D2&C1\r]
Nov 19 16:50:01.50: [20798]: --> [2:OK]
Nov 19 16:50:01.50: [20798]: <-- [14:AT+FCLASS=2.0\r]
Nov 19 16:50:01.54: [20798]: --> [2:OK]
Nov 19 16:50:01.54: [20798]: <-- [9:AT+FLO=1\r]
Nov 19 16:50:01.58: [20798]: --> [2:OK]
Nov 19 16:50:01.58: [20798]: <-- [9:AT+FPP=0\r]
Nov 19 16:50:01.62: [20798]: --> [2:OK]
Nov 19 16:50:01.63: [20798]: <-- [9:AT+FBO=0\r]
Nov 19 16:50:01.66: [20798]: --> [2:OK]
Nov 19 16:50:01.66: [20798]: <-- [10:AT+FCT=30\r]
Nov 19 16:50:01.70: [20798]: --> [2:OK]
Nov 19 16:50:01.70: [20798]: <-- [15:AT+FNR=1,1,1,1\r]
Nov 19 16:50:01.74: [20798]: --> [2:OK]
Nov 19 16:50:01.74: [20798]: <-- [9:AT+FIE=0\r]
Nov 19 16:50:01.78: [20798]: --> [2:OK]
Nov 19 16:50:01.78: [20798]: <-- [23:AT+FCC=1,5,2,2,0,0,0,0\r]
Nov 19 16:50:01.83: [20798]: --> [2:OK]
Nov 19 16:50:01.83: [20798]: <-- [5:ATM0\r]
Nov 19 16:50:01.99: [20798]: --> [2:OK]
Nov 19 16:50:02.04: [20798]: <-- [14:AT+FCLASS=2.0\r]
Nov 19 16:50:02.18: [20798]: --> [2:OK]
Nov 19 16:50:02.18: [20798]: <-- [9:AT+FLO=1\r]
Nov 19 16:50:02.32: [20798]: --> [2:OK]
Nov 19 16:50:02.32: [20798]: <-- [9:AT+FPP=0\r]
Nov 19 16:50:02.46: [20798]: --> [2:OK]
Nov 19 16:50:02.46: [20798]: <-- [9:AT+FBO=0\r]
Nov 19 16:50:02.60: [20798]: --> [2:OK]
Nov 19 16:50:02.60: [20798]: <-- [10:AT+FCT=30\r]
Nov 19 16:50:02.74: [20798]: --> [2:OK]
Nov 19 16:50:02.74: [20798]: <-- [15:AT+FNR=1,1,1,1\r]
Nov 19 16:50:02.88: [20798]: --> [2:OK]
Nov 19 16:50:02.88: [20798]: <-- [9:AT+FIE=0\r]
Nov 19 16:50:03.03: [20798]: --> [2:OK]
Nov 19 16:50:03.03: [20798]: <-- [23:AT+FCC=1,5,2,2,0,0,0,0\r]
Nov 19 16:50:03.17: [20798]: --> [2:OK]
Nov 19 16:50:03.17: [20798]: <-- [22:AT+FLI="NothingSetup"\r]
Nov 19 16:50:03.32: [20798]: --> [2:OK]
Nov 19 16:50:03.33: [20798]: DIAL 3610,8120767246029,0231884129#
Nov 19 16:50:03.33: [20798]: <-- [35:ATDT3610,8120767246029,0231884129#\r]
Nov 19 16:50:37.28: [20798]: --> [4:+FCO]
Nov 19 16:50:38.52: [20798]: --> [13:+FNF:BC530170]
Nov 19 16:50:38.52: [20798]: REMOTE NSF "BC530170"
Nov 19 16:50:39.25: [20798]: --> [27:+FCI:"      02 31 88 41 29"]
Nov 19 16:50:39.25: [20798]: REMOTE CSI "02 31 88 41 29"
Nov 19 16:50:39.54: [20798]: --> [20:+FIS:1,3,0,2,0,0,0,4]
Nov 19 16:51:11.33: [20798]: --> [7:+FHS:11]
Nov 19 16:51:11.33: [20798]: REMOTE HANGUP: No answer (T.30 T1 timeout) (code 
11)
Nov 19 16:51:11.34: [20798]: SEND FAILED: No answer (T.30 T1 timeout)
Nov 19 16:51:11.34: [20798]: SEND FAILED: No answer (T.30 T1 timeout); too 
many attempts to send
Nov 19 16:51:11.34: [20798]: <-- [5:ATH0\r]
Nov 19 16:51:11.34: [20798]: --> [2:OK]
Nov 19 16:51:11.36: [20798]: SESSION END

    ---- Unsent job status ----

   Destination: 3610,8120767246029,0231884129#
         JobID: 4061
       GroupID: 4061
        Sender: nobody
      Mailaddr: nobody@francenet.fr
        CommID: 00006313
Submitted From: localhost.francenet.fr
    Page Width: 209 (mm)
   Page Length: 296 (mm)
    Resolution: 196 (lpi)
        Status: No answer (T.30 T1 timeout); too many attempts to send
       Dialogs: 3 (exchanges with remote device)
         Dials: 0 (consecutive failed calls to destination)
         Calls: 3 (total phone calls placed)
         Pages: 0 (pages transmitted)
      TotPages: 1 (total pages to transmit)
      Attempts: 0 (attempts to send current page)
        Dirnum: 0 (directory of next page to send)

    ---- Documents submitted for transmission ----

The following documents were submitted for transmission and are
available on the server for reuse until they are automatically
purged when this job is removed.  Documents may also be manually
removed using the faxrm command; consult faxrm(1) for information.

Filename                 Size Type
docq/doc4081.ps          7005 PostScript
-- 
Gildas PERROT, perrot@francenet.fr         __o
FranceNet, 28 rue Desaix, 75015 Paris ---_ \<,_
http://www.francenet.fr            ---- (_)/ (_)

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: sendpage
To: tkutt@golden.net (Tim Kutt)
Date: Sat, 22 Nov 1997 08:20:33 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Tim Kutt wrote:
    
    Hello Everyone ! 
    
    I am having a few problems with sendpage, here is the session log, we are
    using Pagenet Canada. 
    
    Any Ideas ? 

The login prologue into Pagenet looks fine and the
syntax of the message too. Please check with Pagenet
why they're not accepting the message. There is a delay
of ~2 secs between pagesend(1M)'s message transmit and
the NAK response of Pagenet, looks like they are checking
something (e.g. to validate the correctness and completeness
of the pager number). But this is only a wild guess.
Ask Pagenet and show them the log.

	matthias

    
    --------------------------
    
    Your pager job to 9,2446161 failed because:
        Protocol failure: paging central responded to message block transmit
    with forced disconnect
    
        ---- Transcript of session follows ----
    
    Nov 21 18:03:21.96: [ 7338]: SESSION BEGIN 00001220 151992446161
    Nov 21 18:03:21.96: [ 7338]: SEND PAGE: JOB 668 DEST 9,2446161 COMMID 00001220
    Nov 21 18:03:21.96: [ 7338]: MODEM set DTR OFF
    Nov 21 18:03:21.96: [ 7338]: DELAY 2600 ms
    Nov 21 18:03:24.57: [ 7338]: MODEM set DTR ON
    Nov 21 18:03:24.57: [ 7338]: MODEM set baud rate: 2400 baud, input flow
    RTS/CTS, output flow RTS/CTS
    Nov 21 18:03:24.57: [ 7338]: MODEM flush i/o
    Nov 21 18:03:24.57: [ 7338]: <-- [15:ATE0V1Q0S0=0H0\r]
    Nov 21 18:03:24.69: [ 7338]: --> [2:OK]
    Nov 21 18:03:24.69: [ 7338]: <-- [21:ATS8=2S7=60&K3&D2&C1\r]
    Nov 21 18:03:24.83: [ 7338]: --> [2:OK]
    Nov 21 18:03:24.83: [ 7338]: <-- [5:ATM0\r]
    Nov 21 18:03:24.90: [ 7338]: --> [2:OK]
    Nov 21 18:03:24.90: [ 7338]: MODEM set parity: 7 bits, even parity
    Nov 21 18:03:24.90: [ 7338]: MODEM input buffering enabled
    Nov 21 18:03:24.90: [ 7338]: <-- [12:AT+FCLASS=0\r]
    Nov 21 18:03:25.10: [ 7338]: --> [2:OK]
    Nov 21 18:03:25.10: [ 7338]: DIAL 9,2446161
    Nov 21 18:03:25.10: [ 7338]: <-- [15:ATDT9,2446161@\r]
    Nov 21 18:03:42.72: [ 7338]: --> [12:CONNECT 2400]
    Nov 21 18:03:42.75: [ 7338]: EXPECT ID (paging central identification)
    Nov 21 18:03:42.75: [ 7338]: <-- <1:\015>
    Nov 21 18:03:44.74: [ 7338]: MODEM TIMEOUT: reading line from modem
    Nov 21 18:03:44.74: [ 7338]: <-- <1:\015>
    Nov 21 18:03:45.05: [ 7338]: --> <3:ID=>
    Nov 21 18:03:45.05: [ 7338]: RECV ID ("ID=")
    Nov 21 18:03:45.06: [ 7338]: SEND device identification/login request
    Nov 21 18:03:45.06: [ 7338]: <-- <5:\033PG1\015>
    Nov 21 18:03:46.64: [ 7338]: --> <1:\006>
    Nov 21 18:03:46.64: [ 7338]: RECV ACK (login successful)
    Nov 21 18:03:46.64: [ 7338]: --> <24:Processing - Please wait>
    Nov 21 18:03:46.65: [ 7338]: RECV: Processing - Please wait
    Nov 21 18:03:46.83: [ 7338]: --> <3:\033[p>
    Nov 21 18:03:46.83: [ 7338]: RECV go-ahead (prologue done)
    Nov 21 18:03:46.84: [ 7338]: SEND message block
    Nov 21 18:03:46.84: [ 7338]: <-- <29:\0025190400920\015test to
    tee\015\00363>\015>
    Nov 21 18:03:48.31: [ 7338]: --> <1:\025>
    Nov 21 18:03:48.31: [ 7338]: RECV NAK (message block rejected)
    Nov 21 18:03:48.31: [ 7338]: SEND message block (retransmit)

	<... log truncated ...>

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Sun, 23 Nov 1997 22:24:21 +0100
From: Ramon Arxer i Galabert <arxer@bcn.servicom.es>
Organization: http://yi.com/home/ArxerRamon/
To: Llista distribucio HylaFAX <flexfax@sgi.com>
CC: Nico Garcia <raoul@cirl.meei.harvard.edu>,
        Matthias Apitz <Matthias.Apitz@SOFTCON.de>
Subject: flexfax: sendfax (the end...?)
Sender: owner-flexfax@celestial.com

First of all thanks to all the folks of the flexfax list.
I finally found the problem of my system when I tried to use sendfax.
An error creating a link to the ps2fax.gs script and bad font files were
the origin of the problem.
-- 
Best regards,

Ramon Arxer i Galabert

e-mail :
	arxer@bcn.servicom.es
	ramon25@casal.upc.es
WWW :
	http://yi.com/home/ArxerRamon
	http://casal.upc.es/~ramon25

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: BUG: ModemMinSpeed config parameter does nothing
To: hahne@aisol.net (Bruce M. Hahne)
Date: Mon, 24 Nov 1997 14:59:01 +0100 (MEZ)
Cc: flexfax@sgi.com, guru@thias.muc.de
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Bruce M. Hahne wrote:
    
    This bug is confirmed under 4.0pl0 and probably exists in all more recent
    releases as well.
    
    Context: Hylafax 4.0pl0 with class 2 modems.  OS should be irrelevant.
    Probably occurs for class 2.0 modems also.
    
    Bug: ModemMinSpeed parameter documented in Config(4f) man page does
    nothing due to incorrect IF test.
    
    There are two ways to specify a minimum baud rate to Hylafax: in the per-tty
    config files and using the -b SPEED command-line switch to sendfax(1).  The
    latter works; the former does nothing.
    
    The reason for the bug appears to be this line in Class2Send.c++, which
    issues the actual minspeed command to the modem if required:
    
    if(req.minsp != BR_2400 && !class2Cmd(minspCmd,req.minsp))
    
    The parameter tested is req.minsp, part of a FaxRequest which is generated
    by the original sendfax command issued on the command line.  Although there
    is a separate FaxModem::minsp parameter which is correctly set to be the max
    of the command-line value and the tty's config file value, that isn't what
    gets tested or used above.  So the config file line is ignored when
    determining whether or not to order the modem to set a minimum speed.
    
    I haven't worked out a fix yet, though it's probably as simple as changing
    "req.minsp" to "minsp" in the IF test above.
    
    Adding the -b SPEED command to my sendfax calls correctly caused Hylafax to
    issue an AT+FMINSP command to the modem prior to sending the fax.
    
    
    Bruce Hahne                                  TEL: 1-408-248-2275
    Access Internet Solutions Inc.               FAX: 1-408-248-2290
    Worldwide Internet fax services              Web: www.aisol.net
    Santa Clara, CA                              Email: hahne@aisol.net

Thanks for the bug report -- I'll look into that;

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Mon, 24 Nov 1997 16:46:13 -0500
To: flexfax@sgi.com
From: Mike Pistone <mpistone@eurekanet.com>
Subject: flexfax: sendpage problems continued
Sender: owner-flexfax@celestial.com

Here is what I get with the IXO debugging

It looks like the ATH0 line right after CONNECT 1200 is the problem.
Why is the program hanging up?
Just for the record,  I can send pages just fine with this modem from
Windows 95 software so it is not the modem....

ANY HELP?  Please??

Nov 24 18:34:11.20: [10165]: SESSION BEGIN 00001282 18006116532
Nov 24 18:34:11.20: [10165]: MODEM set DTR OFF
Nov 24 18:34:13.81: [10165]: MODEM set DTR ON
Nov 24 18:34:13.81: [10165]: MODEM set baud rate: 19200 baud,input 	flow
XON/XOFF, output flow XON/XOFF
Nov 24 18:34:13.81: [10165]: MODEM flush i/o
Nov 24 18:34:13.81: [10165]: <-- [15:ATE0V1Q0S0=0H0\r]
Nov 24 18:34:13.82: [10165]: --> [2:OK]
Nov 24 18:34:13.82: [10165]: <-- [21:ATS8=2S7=60&K4&D2&C1\r]
Nov 24 18:34:13.84: [10165]: --> [2:OK]
Nov 24 18:34:13.84: [10165]: <-- [5:ATM0\r]
Nov 24 18:34:13.85: [10165]: --> [2:OK]
Nov 24 18:34:13.85: [10165]: MODEM set parity: 7 bits, even parity
Nov 24 18:34:13.85: [10165]: MODEM input buffering enabled
Nov 24 18:34:13.92: [10165]: <-- [12:AT+FCLASS=0\r]
Nov 24 18:34:14.03: [10165]: --> [2:OK]
Nov 24 18:34:14.03: [10165]: <-- [28:ATE0B0S37=5N0W2&Q0F4M1L1\N1\r]
Nov 24 18:34:14.15: [10165]: --> [2:OK]
Nov 24 18:34:14.15: [10165]: <-- [17:ATDT18006116532@\r]
Nov 24 18:34:29.11: [10165]: --> [12:CARRIER 1200]
Nov 24 18:34:29.11: [10165]: <-- [5:ATH0\r]
Nov 24 18:34:34.11: [10165]: MODEM TIMEOUT: reading line from modem
Nov 24 18:34:34.11: [10165]: MODEM <Timeout>
Nov 24 18:34:34.11: [10165]: MODEM input buffering enabled
Nov 24 18:34:34.12: [10165]: MODEM set DTR OFF
Nov 24 18:34:34.12: [10165]: SESSION END

-------------------------------------------------------------
Mike Pistone                            pistone@eurekanet.com
Systems/Network Administrator                 ph 614.593-5052
Eureka Networks, Ltd.                         fx 614.594-3632

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: sendpage problems continued
To: mpistone@eurekanet.com (Mike Pistone)
Date: Tue, 25 Nov 1997 08:15:28 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

Mike Pistone wrote:
    
    Here is what I get with the IXO debugging
    
    It looks like the ATH0 line right after CONNECT 1200 is the problem.
    Why is the program hanging up?
    Just for the record,  I can send pages just fine with this modem from
    Windows 95 software so it is not the modem....

BTW: If a modem returns garbage or noice and the Windows 95
sw is designed to eat up the garbage for that specific modem
this does not mean that the modem isn't the problem.
And if your Windows 95 software runs fine with that modem it
would be nice to see a log file/trace *how* this software
handles the problem below.
    
    ANY HELP?  Please??

See below.
    
    Nov 24 18:34:11.20: [10165]: SESSION BEGIN 00001282 18006116532
    Nov 24 18:34:11.20: [10165]: MODEM set DTR OFF
    Nov 24 18:34:13.81: [10165]: MODEM set DTR ON
    Nov 24 18:34:13.81: [10165]: MODEM set baud rate: 19200 baud,input 	flow
    XON/XOFF, output flow XON/XOFF
    Nov 24 18:34:13.81: [10165]: MODEM flush i/o
    Nov 24 18:34:13.81: [10165]: <-- [15:ATE0V1Q0S0=0H0\r]
    Nov 24 18:34:13.82: [10165]: --> [2:OK]
    Nov 24 18:34:13.82: [10165]: <-- [21:ATS8=2S7=60&K4&D2&C1\r]
    Nov 24 18:34:13.84: [10165]: --> [2:OK]
    Nov 24 18:34:13.84: [10165]: <-- [5:ATM0\r]
    Nov 24 18:34:13.85: [10165]: --> [2:OK]
    Nov 24 18:34:13.85: [10165]: MODEM set parity: 7 bits, even parity
    Nov 24 18:34:13.85: [10165]: MODEM input buffering enabled
    Nov 24 18:34:13.92: [10165]: <-- [12:AT+FCLASS=0\r]
    Nov 24 18:34:14.03: [10165]: --> [2:OK]
    Nov 24 18:34:14.03: [10165]: <-- [28:ATE0B0S37=5N0W2&Q0F4M1L1\N1\r]
    Nov 24 18:34:14.15: [10165]: --> [2:OK]
    Nov 24 18:34:14.15: [10165]: <-- [17:ATDT18006116532@\r]
    Nov 24 18:34:29.11: [10165]: --> [12:CARRIER 1200]

HylFAX's pagesend(1M) dials and your modem returns "CARRIER 1200".
It should return "CONNECT" or "CONNECT 1200" or something like
that. The response "CARRIER...." is treated as a failure.

If you can't switch the message off (check your modem manual)
there are now two opportunities. Check (e.g. with a terminal
program) whether or not the "CARRIER...." message is followed
by a "CONNECT" message. In this case you could add the eating
up of the garbage to ClassModem::isNoise() in faxd/ClassModem.c++
The other opportunity is to use another modem for HylaFAX.

	matthias

    Nov 24 18:34:29.11: [10165]: <-- [5:ATH0\r]
    Nov 24 18:34:34.11: [10165]: MODEM TIMEOUT: reading line from modem
    Nov 24 18:34:34.11: [10165]: MODEM <Timeout>
    Nov 24 18:34:34.11: [10165]: MODEM input buffering enabled
    Nov 24 18:34:34.12: [10165]: MODEM set DTR OFF
    Nov 24 18:34:34.12: [10165]: SESSION END
    
    -------------------------------------------------------------
    Mike Pistone                            pistone@eurekanet.com
    Systems/Network Administrator                 ph 614.593-5052
    Eureka Networks, Ltd.                         fx 614.594-3632

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: "Gary Ng (Winclient)" <garyn@winclient.com>
To: flexfax@sgi.com, "'Mike Pistone'" <mpistone@eurekanet.com>
Subject: RE: flexfax: sendpage problems continued
Date: Tue, 25 Nov 1997 11:31:30 +0800
Sender: owner-flexfax@celestial.com

Just curious, are you sending fax ? AT+FCLASS=0
is for data.

Gary

> ----------
> From: 	Mike Pistone[SMTP:mpistone@eurekanet.com]
> Sent: 	Tuesday, November 25, 1997 5:46 AM
> To: 	flexfax@sgi.com
> Subject: 	flexfax: sendpage problems continued
> 
> Here is what I get with the IXO debugging
> 
> It looks like the ATH0 line right after CONNECT 1200 is the problem.
> Why is the program hanging up?
> Just for the record,  I can send pages just fine with this modem from
> Windows 95 software so it is not the modem....
> 
> ANY HELP?  Please??
> 
> Nov 24 18:34:11.20: [10165]: SESSION BEGIN 00001282 18006116532
> Nov 24 18:34:11.20: [10165]: MODEM set DTR OFF
> Nov 24 18:34:13.81: [10165]: MODEM set DTR ON
> Nov 24 18:34:13.81: [10165]: MODEM set baud rate: 19200 baud,input
> flow
> XON/XOFF, output flow XON/XOFF
> Nov 24 18:34:13.81: [10165]: MODEM flush i/o
> Nov 24 18:34:13.81: [10165]: <-- [15:ATE0V1Q0S0=0H0\r]
> Nov 24 18:34:13.82: [10165]: --> [2:OK]
> Nov 24 18:34:13.82: [10165]: <-- [21:ATS8=2S7=60&K4&D2&C1\r]
> Nov 24 18:34:13.84: [10165]: --> [2:OK]
> Nov 24 18:34:13.84: [10165]: <-- [5:ATM0\r]
> Nov 24 18:34:13.85: [10165]: --> [2:OK]
> Nov 24 18:34:13.85: [10165]: MODEM set parity: 7 bits, even parity
> Nov 24 18:34:13.85: [10165]: MODEM input buffering enabled
> Nov 24 18:34:13.92: [10165]: <-- [12:AT+FCLASS=0\r]
> Nov 24 18:34:14.03: [10165]: --> [2:OK]
> Nov 24 18:34:14.03: [10165]: <-- [28:ATE0B0S37=5N0W2&Q0F4M1L1\N1\r]
> Nov 24 18:34:14.15: [10165]: --> [2:OK]
> Nov 24 18:34:14.15: [10165]: <-- [17:ATDT18006116532@\r]
> Nov 24 18:34:29.11: [10165]: --> [12:CARRIER 1200]
> Nov 24 18:34:29.11: [10165]: <-- [5:ATH0\r]
> Nov 24 18:34:34.11: [10165]: MODEM TIMEOUT: reading line from modem
> Nov 24 18:34:34.11: [10165]: MODEM <Timeout>
> Nov 24 18:34:34.11: [10165]: MODEM input buffering enabled
> Nov 24 18:34:34.12: [10165]: MODEM set DTR OFF
> Nov 24 18:34:34.12: [10165]: SESSION END
> 
> -------------------------------------------------------------
> Mike Pistone                            pistone@eurekanet.com
> Systems/Network Administrator                 ph 614.593-5052
> Eureka Networks, Ltd.                         fx 614.594-3632
> 

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: sendpage problems continued
To: mpistone@eurekanet.com (Mike Pistone)
Date: Tue, 25 Nov 1997 08:31:58 +0000 (GMT)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> 
> Here is what I get with the IXO debugging
> 
> It looks like the ATH0 line right after CONNECT 1200 is the problem.
                                          ^^^^^^^
You are not getting CONNECT (you are getting CARRIER).  However I don't
understand why it is responding to this with a software hangup.

> Nov 24 18:34:29.11: [10165]: --> [12:CARRIER 1200]
> Nov 24 18:34:29.11: [10165]: <-- [5:ATH0\r]
> Nov 24 18:34:34.11: [10165]: MODEM TIMEOUT: reading line from modem

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: jared@4websites.com
Date: Tue, 25 Nov 1997 17:23:10 -0800
To: flexfax@sgi.com
Subject: flexfax: Hylafax
Sender: owner-flexfax@celestial.com

Hylafax v4.0pl1 under BSDI 2.1 with gcc 2.7.2 using Hayes 14.4 & Hayes
28.8

I have just recently started having a problem that has the symptoms of
the documents in the docq directory being over writen.
Example:
	Bob sends a request for a new car to 555-1212 with an ATTN: Simon
	Fred sends a request for a travel reservation to 555-1313 with an ATTN:
Sally
	The request for the new car gets sent to 555-1313 with the proper ATTN:

We send nearly 600 faxes per day. This problem does not occur very
often, however.


Thank you,
Jared Tracy

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: Matthias.Apitz@SOFTCON.de
Subject: Re: flexfax: Hylafax
To: jared@4websites.com
Date: Wed, 26 Nov 1997 08:31:24 +0100 (MEZ)
Cc: flexfax@sgi.com
>From: guru@sisis.de (Matthias Apitz)
Reply-To: Matthias.Apitz@SOFTCON.de (Matthias Apitz)
X-FAX-cover: faxcover-sisis.ps
Sender: owner-flexfax@celestial.com

jared@4websites.com wrote:
    
    Hylafax v4.0pl1 under BSDI 2.1 with gcc 2.7.2 using Hayes 14.4 & Hayes
    28.8
    
    I have just recently started having a problem that has the symptoms of
    the documents in the docq directory being over writen.
    Example:
    	Bob sends a request for a new car to 555-1212 with an ATTN: Simon
    	Fred sends a request for a travel reservation to 555-1313 with an ATTN:
    Sally
    	The request for the new car gets sent to 555-1313 with the proper ATTN:
    
    We send nearly 600 faxes per day. This problem does not occur very
    often, however.

There is a patch of Phil Abercrombie (sent to the list on Sep 25)
fixing a bug in wrapping sequence numbers for jobs and docs.
This bug may cause this problem. Please pick it up from the
archives or come back to me.

If this doesn't solve the problem try to catch such example, e.g
catch the q#### files from sendq or doneq and the referenced
docs.

	matthias

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 27 Nov 1997 23:01:27 +0700 (ICT)
From: Apiset Tananchai <aet@demo.ksc.co.th>
To: flexfax@sgi.com
Subject: flexfax: Problem receiving fax with ZyXEL Elite 2864
Sender: owner-flexfax@celestial.com

Hi all,

I have problem receiving fax using HylaFax4.0 with ZyXEL Elite 2864 on my
Linux 2.0.32 (RedHat 4.2). There was no problem if the sender send only 1
page of fax. When try to send > 1 page I always got "Unspecified Phase C
error, including too much delay between TCF and +FDR command." Follow is
what in the log file:

Nov 27 22:37:18.28: [  726]: RECV: 9907 bytes of data, 2303 total lines
Nov 27 22:37:18.28: [  726]: --> [16:+FPS:1,8FA,0,0,0]
Nov 27 22:37:19.59: [  726]: --> [6:+FET:0]
Nov 27 22:37:19.59: [  726]: RECV recv MPS (more pages, same document)
Nov 27 22:37:19.59: [  726]: --> [2:OK]
Nov 27 22:37:19.64: [  726]: RECV send MCF (message confirmation)
Nov 27 22:37:19.64: [  726]: RECV FAX (00000049): from , page 1 in 2:01, INF, 7.7 line/mm, 1-D MR
Nov 27 22:37:19.64: [  726]: <-- [7:AT+FDR\r]
Nov 27 22:37:34.49: [  726]: --> [7:+FHS:90]
Nov 27 22:37:34.51: [  726]: REMOTE HANGUP: Unspecified Phase C error, including too much delay between TCF and +FDR command (code 90)
Nov 27 22:37:34.55: [  726]: RECV FAX (00000049): recvq/fax00011.tif from, route to <unspecified>, 1 pages in 2:18
Nov 27 22:37:34.60: [  726]: RECV FAX: Unspecified Phase C error, including too much delay between TCF and +FDR command
Nov 27 22:37:34.60: [  726]: RECV FAX (00000049): session with  terminated abnormally: Unspecified Phase C error, including too much delay between TCF and +FDR command
Nov 27 22:37:34.60: [  726]: RECV FAX: bin/faxrcvd "recvq/fax00011.tif" "cua1" "00000049" "Unspecified Phase C error, including too much delay between TCF and +FDR command"
Nov 27 22:37:36.70: [  726]: RECV FAX: end
Nov 27 22:37:36.70: [  726]: SESSION END

What configuration parameter do I have to change in order to reduce delay
between TCF and +FDR command? I can't seem to find in config(5F) man page.

Thanks,
--
aet
.x.

PS: I'm not on the list so please cc'd your reply to my address
    aet@demo.ksc.co.th

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: Problem receiving fax with ZyXEL Elite 2864
To: aet@demo.ksc.co.th (Apiset Tananchai)
Date: Sat, 29 Nov 1997 11:21:06 +0000 (GMT)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> Nov 27 22:37:19.59: [  726]: --> [2:OK]
> Nov 27 22:37:19.64: [  726]: RECV send MCF (message confirmation)
> Nov 27 22:37:19.64: [  726]: RECV FAX (00000049): from , page 1 in 2:01, INF, 7.7 line/mm, 1-D MR
> Nov 27 22:37:19.64: [  726]: <-- [7:AT+FDR\r]
> Nov 27 22:37:34.49: [  726]: --> [7:+FHS:90]
> Nov 27 22:37:34.51: [  726]: REMOTE HANGUP: Unspecified Phase C error, including too much delay between TCF and +FDR command (code 90)
> 
> What configuration parameter do I have to change in order to reduce delay
> between TCF and +FDR command? I can't seem to find in config(5F) man page.

Why is everything fixable by modifying the configuration?  I wouldn't
expect any configuration pararmeter to control the delay.  If there
were a delay the problem would be in the serial driver or underspecified
hardware.  However, you have sent FDR promptly, so the real question is
why did the modem fail to see it (the timeout appears to be 15 seconds,
so this is hardly a performance problem, unless Hylafax is TOO fast for
the modem).

I do have some evidence that the standard Linux serial driver can lose 
characters on some hardware, although this tends to affect a receive that
overlaps a pause in output.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Tue, 25 Nov 1997 13:14:14 +0100
To: flexfax@sgi.com
From: Johnny Oestergaard <joe@agora.dk>
Subject: flexfax: mail-to-fax: coversheet change
Sender: owner-flexfax@celestial.com

Hi

I have a HylaFAX server running as a email-to-fax gateway.

It works fine, but I would like to know how I can change the fax coversheet.

What I need is a way to implement different coversheets for different users
in the /etc/aliases file.

Regards,
Johnny

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 28 Nov 1997 01:39:15 +0100
From: Christoph Lange <clange@jane.biologie.fu-berlin.de>
To: flexfax@sgi.com
Subject: flexfax: Receive time reported by faxstat
X-Organization: Institut fuer Verhaltensbiologie, FU Berlin
X-Priv-Address: Haehnelstr. 14, 12159 Berlin
X-Priv-Telephone: +49 30 859 35 33
Sender: owner-flexfax@celestial.com


Hello, all!

Using Hylafax v4.0pl1 on my Linux 2.0.28 box I have a peculiar problem
(just annoying, anyway, not really destructing service ;-): 'faxstat
-r' shows the GMT of the receive time of faxes. But the file date of
these faxes is local time (one hour ahead of GMT). - Does anyone know if this
is a 'feature' or can I change this behaviour?

	Yours, Christoph.
                        _
--------------------- <(o)
Christoph Lange         T ___
Haehnelstr. 14          |/   \
12159 Berlin            \____/   Phone/Fax : +49 30 8593533
GERMANY                   / |    E-Mail:     clange@biologie.fu-berlin.de
------------------------ - - ---------------------------------------------

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 28 Nov 1997 08:41:33 -0500
From: Yves Trudeau <ytrudeau@jyrmatec.com>
Reply-To: ytrudeau@jyrmatec.com
Organization: Jyrmatec Inc.
To: Christoph Lange <clange@jane.biologie.fu-berlin.de>
CC: flexfax@sgi.com
Subject: Re: flexfax: Receive time reported by faxstat
Sender: owner-flexfax@celestial.com

Christoph Lange wrote:
> 
> Hello, all!
> 
> Using Hylafax v4.0pl1 on my Linux 2.0.28 box I have a peculiar problem
> (just annoying, anyway, not really destructing service ;-): 'faxstat
> -r' shows the GMT of the receive time of faxes. But the file date of
> these faxes is local time (one hour ahead of GMT). - Does anyone know if this
> is a 'feature' or can I change this behaviour?
> 
>         Yours, Christoph.

You could try 'faxstat -lr'. The 'l' stand for local time.
-- 

Yves Trudeau, Ph. D.
Jyrmatec Inc.
1381, Jolliet
Sherbrooke, Quebec
Canada, J1K 1Y8
Tel.:(819) 565-8349
Fax.:(819) 565-8273
email: ytrudeau@jyrmatec.com

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: Receive time reported by faxstat
To: clange@jane.biologie.fu-berlin.de (Christoph Lange)
Date: Fri, 28 Nov 1997 08:43:57 +0000 (GMT)
Cc: flexfax@sgi.com
Sender: owner-flexfax@celestial.com

> -r' shows the GMT of the receive time of faxes. But the file date of
> these faxes is local time (one hour ahead of GMT). - Does anyone know if this
> is a 'feature' or can I change this behaviour?

Assuming you mean the time displayed by ls -l (file dates should ALWAYS be
UTC), you can override the timezone for ls.  E.g. for sh or bash:

TZ=GMT ls -l

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Thu, 27 Nov 1997 23:00:54 -0800
From: Steve Pershing <sgp@bc.sympatico.ca>
Organization: Questor Technologies Inc.
To: flexfax@sgi.com
Subject: flexfax: Hylafax 4.0pl1 doesn't recognize distinctive ring for fax/voice
Sender: owner-flexfax@celestial.com

I'm running Linux 2.1.66 on a 486 box, and have not been able to get
Hylafax to recognise the "RING 1" return code from my ZyXEL 1496S+ (same
as E+) modem.

I'd like it to recognize "RING" for voice messages, and "RING 1" for
incoming faxes.

Also, I'd sure like to be able to enable it to work in voice mode.

Has anyone had success in doing this, and if so how was it implemented??

Thanks in advance for your responses.


--
  Steve Pershing

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Sat, 29 Nov 1997 22:58:10 +0100
To: flexfax@sgi.com
From: Karl Feichtinger <lists@comconnect.com>
Subject: flexfax: Using Hylafax with Winword mail merge 
Cc: lorenzo@argon.roma2.infn.it
Sender: owner-flexfax@celestial.com

I tried to post on this subject before but took a wrong translation of the
"Seriendruck" term of my German Winword version. Im sorry for having
caused a misunderstanding.

This is a question related to connecting Winword to a HylaFax server:
I want to use the Winword mail merge feature in order to send multiple
faxes in one run. WHFC seems not to be a solution for this, since this can
only send ONE fax at a time. I have a vage idea how this could be done,
however, one step is missing:

The Winword mail merge feature can send emails with an DOC-type file
attachment to email addresses taken from the mail merge database table. I
could encode the fax number in the email address. The unix side mail server
then could decode the fax number and pass the file attachment on to
HylaFax. However, the DOC must be translated to PS or TIFF first. 

Is there a know solution for this? 
Or is there a totally different approach to solve this application?

Any help would be greatly appreciated!

--Karl

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
From: David Woolley <david@djwhome.demon.co.uk>
Subject: Re: flexfax: Using Hylafax with Winword mail merge
To: lists@comconnect.com (Karl Feichtinger)
Date: Sun, 30 Nov 1997 11:58:42 +0000 (GMT)
Cc: flexfax@sgi.com, lorenzo@argon.roma2.infn.it
Sender: owner-flexfax@celestial.com

> The Winword mail merge feature can send emails with an DOC-type file
> attachment to email addresses taken from the mail merge database table. I
> could encode the fax number in the email address. The unix side mail server
> then could decode the fax number and pass the file attachment on to
> HylaFax. However, the DOC must be translated to PS or TIFF first.=20
> 
> Is there a know solution for this?=20
> Or is there a totally different approach to solve this application?

This is a second level FAQ.  Microsoft have very little interest in
supporting Unix or in batch mode programs.  A freeware implementor
would probably find this a major undertaking, assuming that they could
obtain the file format information for Word.  Nobody has yet reported
a commercial option, and it is probably the case that there is an
insufficient market, or it is too difficult to track the file format
(unless Microsoft cooperates, it is always going to be several months
behind the latest Microsoft Word format).

I know it is a hopeless task convincing people of this, but for a large
mailmerge run, it probably doesn't take long to prepare a troff master
pattern and use Perl to do the substitution, as maintaining the list
will be the main cost.

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Fri, 28 Nov 1997 16:47:37 +0100
To: flexfax@sgi.com
From: Karl Feichtinger <lists@comconnect.com>
Subject: flexfax: Individual mass faxes from Winword via Hylafax
Sender: owner-flexfax@celestial.com

I have a request for a mass faxing application which I would like to
implement with Hylafax on a Linux machine, and wonder if anyone has any
advice or practical experience:

I need to send mass mailings by fax, one page for each receiver. The faxes
are not not identical, but address each receiver individually by name etc.
This is my idea:

With MS Winwords mass-mailing-via-e-mail option the client side would send
e-mail messages, and a mail server (sendmail) could transform this into
faxes (with the fax number taken from the e-mail address). The problem is:
the e-mail would contain an MS Word document file attachment which Hylafax
cannot transform into TIFF. I would need postscript instead.

The WHFC Windows 95 Hylafax client seems not to be an option in this case
since WHFC can only deliver one fax at a time, with manual typing in of the
fax number.

Any help would be greatly appreciated!

	Karl Feichtinger

From owner-flexfax@celestial.com  Wed Feb 17 13:55:48 1999
Date: Sun, 30 Nov 1997 16:46:45 -0700 (MST)
From: Zach Tippit <sterno@verinet.com>
To: flexfax@sgi.com
Subject: flexfax: Modem reset troubles
Sender: owner-flexfax@celestial.com



It seems that either hfaxd or faxsend does not reset my modem after
sending a  fax; If I send one it goes fine, but the next one will not work
until I do "echo "atz" > /dev/modem" and then force the queue.  Is there
any way to make this happen automagically after sending a fax?

Zach

