Printer Problems / Mail Msgs/ HP Jetdirect (External)
Dave Gresham
several HP External Jet Direct Cards that allow one to plug a
parllel printer in either serial or parallel.
A couple of months ago I was called in to find out why their system
was going so slow, and realized that they were running out of disk
space. I found out that the reason they were running out of space
was that the lp system was sending huge email messages of up to
8000 lines or more, and eventually some users had mailboxes up
to 80mb or more. And since they do not use their system presently
for email it just kept on growing.
i removed all the mailboxes, but they keep sending these messages, so
i am hoping someone might be able to give me an idea or direction on
how to correct this. the messages they are getting and this
message can be repeated hundreds, if not thousands of times in
a single mail message. The message is:
Subject: Status of lp request srlaser9-90173
Your request srlaser9-90173 destined for srlaser9
encountered an error while printing on printer srlaser9.
Reason for failure:
The output ``port'', a FIFO, was closed before all output was written.
The output ``port'', a FIFO, was closed before all output was written.
The output ``port'', a FIFO, was closed before all output was written.
The output ``port'', a FIFO, was closed before all output was written.
The output ``port'', a FIFO, was closed before all output was written.
The output ``port'', a FIFO, was closed before all output was written.
I am using the hpnp. script, here is the first few lines from
/usr/spool/lp/admins/interfaces/srlaser9 script
#!/bin/sh
#
# @(#) hpnp.model 25.2 95/02/27
#
# $Header: hpnp.model,v 1.9 91/11/14 10:44:35 pma Exp $
#
I have it going to port 9101
any ideas, or if i can proviide more infomration i will.
dave
Steve Wertz
: Reason for failure:
: The output ``port'', a FIFO, was closed before all output was written.
: The output ``port'', a FIFO, was closed before all output was written.
: The output ``port'', a FIFO, was closed before all output was written.
: The output ``port'', a FIFO, was closed before all output was written.
: The output ``port'', a FIFO, was closed before all output was written.
: The output ``port'', a FIFO, was closed before all output was written.
: I am using the hpnp. script, here is the first few lines from
: /usr/spool/lp/admins/interfaces/srlaser9 script
: #!/bin/sh
: #
: # @(#) hpnp.model 25.2 95/02/27
: #
: # $Header: hpnp.model,v 1.9 91/11/14 10:44:35 pma Exp $
: #
: I have it going to port 9101
This really isn't the interface script. It's a wrapper for the interface
script. The real interafce is down in ./model.orig/<printer-name>...
Use the /usr/spool/lp/model/HPLaserJet script if its an HP Laserjet
printer, and also make sure you have a live/functional printer
on port 9100 before you try and use 9101 (or 9102).
Still no-go? Then post the printer model, print server model, and
your interface script as it appears in:
/usr/spool/lp/admins/lp/interfaces/model.orig
-sw
Jeff Liebermann
>I have a client on SCO Openserver 5.0.2 Enterprise and they have
>several HP External Jet Direct Cards that allow one to plug a
>parllel printer in either serial or parallel.
Ahah. A detective story. I deduce by the fact that you used the term
"external" that the probability that you unspecified HP Jet Direct is not a
card but a print server with a high probability of being an HP EX3, 300x,
EX Plus, 170X, or 500X print server. Since you've also leaked the hint
that it can be either a parallel or serial connection, I would deduce from
the above information that it is probably not an HP print server as none of
the ones in my catalog have a single parallel and single serial port. I
realize that supplying a model number is extermely difficult, but it does
help identify the problem.
My guess is that you actually have an EX3 print server based upon the
apparent age of the installation. This was the first HP external print
server and there are some nasty problems. Ancient versions of the firmware
would sometimes refuse to print the last page. I don't recall other
firmware problems, but checking the HP web pile for the latest flash ram
version of your unspecified printer might be helpful.
To insure that your life will be entertaining, the SCO supplied HPNP
printing utilities have a nasty bug. HPNP uses SNMP (Stupid Network
Management Polluter) to determine if the print server is alive and ready to
accept jobs. However, it only checks port #1. If you take the printer
plugged into port #1 offline, the print server will not print to the other
two ports. I vaguely recall messages similar to what you were getting when
this happened.
The easiest way to troubleshoot HP JetDirect problems is to fire up a
Windoze box with HP JetAdmin and monitor what the print server is doing.
JetAdmin is basically an SNMP monitor that is hard coded for the JetDirect
boxes. You could do the same with an SNMP MIB browser, but you'll be
buried in alarms and traps.
There is a small chance that if your unspecified print server really is an
EX3, you may have it improperly configured. Port 9101 is the port number
for the 2nd printer. The three port numbers are 9100, 9101 and 9102.
Perhaps you're printing to the wrong port number?
How do I configure the HP JetDirect EX Plus3 Print Server to use HPNP?
http://www.sco.com/cgi-bin/ssl_reference?105327
How do I stop hpnp logfiles from accumulating in my /tmp directory?
http://www.sco.com/cgi-bin/ssl_reference?105883
Just for reference, the EX3 will also work with lpd/lpr. See:
http://www.sco.com/cgi-bin/ssl_reference?104997
There are other articles. Plug the keyword "HPNP" into:
http://www.sco.com/ta/
and RTFM.
--
Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
(831)421-6491 pgr (831)426-1240 fax (831)336-2558 home
http://www.cruzio.com/~jeffl WB6SSY
je...@comix.santa-cruz.ca.us je...@cruzio.com
Dave Gresham
>Dave Gresham <de...@netcom.com> wrote:
>
>: #
>: # $Header: hpnp.model,v 1.9 91/11/14 10:44:35 pma Exp $
>: #
>
>: I have it going to port 9101
>
>This really isn't the interface script. It's a wrapper for the interface
>script. The real interafce is down in ./model.orig/<printer-name>...
>Use the /usr/spool/lp/model/HPLaserJet script if its an HP Laserjet
>printer, and also make sure you have a live/functional printer
>on port 9100 before you try and use 9101 (or 9102).
>
>Still no-go? Then post the printer model, print server model, and
>your interface script as it appears in:
>/usr/spool/lp/admins/lp/interfaces/model.orig
>
>-sw
Sorry for not supplying all the information, it was a friday and I got sloppy.
I am using the HPLaserJet script and this is an EX Plus 3 Print server.
this printer is indeed on the 2nd port, and I already have a working
printer on the first port. There are jobs printing ok to this 2nd port, its
that we are getting these email messages and I'm trying to figure out
what might be causing them.
Dave Gresham
Jeff Liebermann <je...@comix.santa-cruz.ca.us> wrote:
>On Thu, 1 Apr 1999 16:40:30 GMT, de...@netcom.com (Dave Gresham) wrote:
>
>>I have a client on SCO Openserver 5.0.2 Enterprise and they have
>>several HP External Jet Direct Cards that allow one to plug a
>>parllel printer in either serial or parallel.
>
>apparent age of the installation. This was the first HP external print
>server and there are some nasty problems. Ancient versions of the firmware
>would sometimes refuse to print the last page. I don't recall other
>firmware problems, but checking the HP web pile for the latest flash ram
>version of your unspecified printer might be helpful.
>
Sorry, it was a friday or so when I wrote that message and should of
provided better information. It is an EX Plus 3.
>To insure that your life will be entertaining, the SCO supplied HPNP
>printing utilities have a nasty bug. HPNP uses SNMP (Stupid Network
>Management Polluter) to determine if the print server is alive and ready to
>accept jobs. However, it only checks port #1. If you take the printer
>plugged into port #1 offline, the print server will not print to the other
>two ports. I vaguely recall messages similar to what you were getting when
>this happened.
>
>The easiest way to troubleshoot HP JetDirect problems is to fire up a
>Windoze box with HP JetAdmin and monitor what the print server is doing.
>JetAdmin is basically an SNMP monitor that is hard coded for the JetDirect
>boxes. You could do the same with an SNMP MIB browser, but you'll be
>buried in alarms and traps.
>
>There is a small chance that if your unspecified print server really is an
>EX3, you may have it improperly configured. Port 9101 is the port number
>for the 2nd printer. The three port numbers are 9100, 9101 and 9102.
>Perhaps you're printing to the wrong port number?
I already have a working printer on the first port, and this printer
in indeed on the 2nd port. What is interesting is that print jobs are
sucessfully going to this printer, its only that there are times when we
get these huge email messages. Maybe it is because of the printer being
offline, and in the past year I have upgraded the firmware on the
EXPlus 3 in the past year when there was even nastier bugs in it.
>How do I configure the HP JetDirect EX Plus3 Print Server to use HPNP?
> http://www.sco.com/cgi-bin/ssl_reference?105327
Already have this.
>How do I stop hpnp logfiles from accumulating in my /tmp directory?
> http://www.sco.com/cgi-bin/ssl_reference?105883
>
Already applied this fix.
Steve Wertz
: In article <7eb93f$p...@avon.x.co.uk>, Steve Wertz <swe...@sco.com> wrote:
:>Dave Gresham <de...@netcom.com> wrote:
:>
:>: #
:>: # $Header: hpnp.model,v 1.9 91/11/14 10:44:35 pma Exp $
:>: #
:>
:>: I have it going to port 9101
:>
:>This really isn't the interface script. It's a wrapper for the interface
:>script. The real interafce is down in ./model.orig/<printer-name>...
:>Use the /usr/spool/lp/model/HPLaserJet script if its an HP Laserjet
:>printer, and also make sure you have a live/functional printer
:>on port 9100 before you try and use 9101 (or 9102).
:>
:>Still no-go? Then post the printer model, print server model, and
:>your interface script as it appears in:
:>/usr/spool/lp/admins/lp/interfaces/model.orig
:>
:>-sw
: Sorry for not supplying all the information, it was a friday and I got sloppy.
: I am using the HPLaserJet script and this is an EX Plus 3 Print server.
Has it been modified? diff it with /usr/spool/lp/model/HPlaserJet.
: this printer is indeed on the 2nd port, and I already have a working
: printer on the first port. There are jobs printing ok to this 2nd port, its
: that we are getting these email messages and I'm trying to figure out
: what might be causing them.
Go with Jeff's suggestion of checking out the BIOS too. You can use
'getmany/getone', or print setup page(s) using the button on the front of
the box. This will give you the BIOS/firmware release.
This shouldn't be happening with the stock HPLaserJet model and
the hpnp.model wrapper with a HP BIOS of 4.00 or better (I'm not
sure if you can flash the HP3EX... never tried it?)
Some of the networking patches for 5.0.2 may help this problem also.
-sw
Bob Willey
HP LX3 Server connected to a 100BaseT connection to an HP switch, rest of
the 16 ports are 10BaseT
with about 8 WIN98 workstations and 4 Print Servers.
2 are internal cards in HP 4000's
2 are external HP 170 Print Servers
The UNIX is 3.2v5.0.5 Enterprise
One of the HP 4000's (first one defined in the bootptab file) prints fine,
all printers can be pinged with no problems, all submasks are 255.255.255.0
PC's are 192.168.10.10 thru .20
UNIX is at 192.168.10.100
Printers are 192.168.10.200 .201 .202 & .203
.200 prints fine, none of the rest.
All PC's can access the UNIX host via telnet with no problems, and can
print to all printers via HP JetAdmin.
On the UNIX side, all configs were done via scoadmin, and HP Printer Manager
Setup for bootp (altered inetd.conf for bootp), and added to the spooler with
2 of the HP 4000's using the HPLaserJet interface
and 2 dot matrix printers using standard
Only the first printer prints fine, none of the rest print, but can be pinged.
Any ideas?
-> Bob Willey - b...@ccs.com <-
CCS Enterprises, Inc.
PO Box 35 Linkwood MD 21835
http://www.ccs.com Come Visit Us !!
http://www.global-homebiz.com/bwilley.html
http://www.intercom.net/user/consult
-> (410) 228-9211 -- FAX: (410) 901-1105 <-
Jeff Liebermann
>I already have a working printer on the first port, and this printer
>in indeed on the 2nd port. What is interesting is that print jobs are
>sucessfully going to this printer, its only that there are times when we
>get these huge email messages.
Ok, let me try to decode this cryptic description.
1. srlaser9 points to port 9101 or port #2 on the EX3 print server.
2. The error messages occur only sometimes with no obvious pattern.
3. The error messages ONLY occur with the printer on port #2.
4. The only email error messages appear from srlaser9. All the other boxes
and ports apparently work normally and don't complain via email.
Is this correct?
>Maybe it is because of the printer being
>offline, and in the past year I have upgraded the firmware on the
>EXPlus 3 in the past year when there was even nastier bugs in it.
If you flashed it within the last year, it's probably current.
If you have SNMP running, invoke:
getmany printer_ip public iso
and you will buried in MIB variables and values. The firmware version
should be on the top line under sysDescr.0. I forgot the exact OID number
or I would have supplied a "getone" command that only returns the desired
variable. For instructions on how to make SNMP work on OSR5, see:
http://www.cruzio.com/~jeffl/sco/snmp_install.txt
Try it. You might like it.
>>How do I configure the HP JetDirect EX Plus3 Print Server to use HPNP?
>> http://www.sco.com/cgi-bin/ssl_reference?105327
>Already have this.
Good. Now check that the values imbedded in the spooler scripts match
those in the Tech Article. It's rather odd that only port #2 *SEEMS* to
having problems. My guess is that either the port is fried on the EX3 in
question, of there is a typo error in the script where the 9101 port number
is defined. Watch out for imbeded backspaces and invisible control
characters. Since you apparently have other EX3 print servers available,
swapping boxes and reconfiguring the IP's will help point the finger to
either the box or the Unix configuration.
Remember: Assumption is the mother of all screwups.
Ken O'Neil
Which one is your default?
What syntax are you using with the lp command to print to the other printers?
Bob Willey
>On Tue, 6 Apr 1999 21:54:45 GMT, in comp.unix.sco.misc you wrote:
>
>
>>Only the first printer prints fine, none of the rest print, but can be
pinged.
>>Any ideas?
>
>1. Run:
> /usr/lib/accept printer_name
> enable printer_name
all were accepted and enabled.
>2. Run HP JetAdmin from one of the DOS boxes and see what it has to say
>about the printer configurations.
The JetAdmin shows the configs basically the same. Nothing that seems out of
wack.
It is the 2nd entry(billing) in the bootptab that is working.
billing & admin are HP 4000's with internal Jetdirect cards
billdot & reception are NEC dot matrix printers with HP 170 Print Servers
Here is the pertinent files:
lpstat -t output:
scheduler is running
system default destination: laser
device for laser: /dev/lp0
device for admin: /dev/null
device for billing: /dev/null
device for billdot: /dev/null
device for reception: /dev/null
laser accepting requests since Tue Apr 06 01:12:49 1999
admin accepting requests since Tue Apr 06 13:54:22 1999
billing accepting requests since Tue Apr 06 13:55:39 1999
billdot accepting requests since Tue Apr 06 13:57:07 1999
reception accepting requests since Tue Apr 06 14:27:14 1999
printer laser is idle. enabled since Tue Apr 06 01:12:49 1999. available.
printer admin is idle. enabled since Tue Apr 06 13:54:22 1999. available.
printer billing is idle. enabled since Tue Apr 06 13:55:39 1999. available.
printer billdot is idle. enabled since Tue Apr 06 13:57:07 1999. available.
printer reception is idle. enabled since Tue Apr 06 14:27:14 1999. available.
/etc/hosts contents:
# @(#)hosts,v 6.1 1993/08/21 02:17:48 stevea Exp - STREAMware TCP/IP
source
# SCCS IDENTIFICATION
127.0.0.1 localhost
192.168.10.100 penneuro penneuro
192.168.10.200 admin
192.168.10.201 billing
192.168.10.202 billdot
192.168.10.203 reception
/etc/bootptab contents:
# @(#)bootptab,v 6.4 1995/01/27 12:54:50
# /etc/bootptab: database for bootp server (/etc/bootpd)
admin:\
:ht=ether:\
:ha=0060B00B70C9:\
:sm=255.255.255.0:\
:hn:\
:ip=192.168.10.200:\
:vm=rfc1048:
billing:\
:ht=ether:\
:ha=0060B004A769:\
:sm=255.255.255.0:\
:hn:\
:ip=192.168.10.201:\
:vm=rfc1048:
billdot:\
:ht=ether:\
:ha=0010830C55A8:\
:sm=255.255.255.0:\
:hn:\
:ip=192.168.10.202:\
:vm=rfc1048:
reception:\
:ht=ether:\
:ha=001083089c73:\
:sm=255.255.255.0:\
:hn:\
:ip=192.168.10.203:\
:vm=rfc1048:
/etc/inetd.conf contents:
# @(#)$Id: inetd.conf,v 6.8 1996/01/09 21:48:54 aes Exp $ - STREAMware
TCP/IP source
#
# Copyrighted as an unpublished work.
# (c) Copyright 1987-1994 Legent Corporation
# All rights reserved.
#
# SCCS IDENTIFICATION
ftp stream tcp nowait root /etc/ftpd ftpd
telnet stream tcp nowait NOLUID /etc/telnetd telnetd
shell stream tcp nowait NOLUID /etc/rshd rshd
login stream tcp nowait NOLUID /etc/rlogind rlogind
exec stream tcp nowait NOLUID /etc/rexecd rexecd
finger stream tcp nowait nouser /etc/fingerd fingerd
#uucp stream tcp nowait NOLUID /usr/lib/uucp/uucpd uucpd
# Enabling this allows public read files to be accessed via TFTP.
tftp dgram udp wait nouser /etc/tftpd tftpd
# This is the more secure method, since only files from /tftpboot can
# be accessed via TFTP. This must be root in order to do the chroot
# to /tftpboot. /tftpboot must be created by hand.
#tftp dgram udp wait root /etc/tftpd tftpd -s /tftpboot
comsat dgram udp wait root /etc/comsat comsat
ntalk dgram udp wait nouser /etc/talkd talkd
#
# Entries for BOOTP and DHCP servers & relay agent
#
# If running tftpd in secure mode, use bootpd with "-c securedir"
# where securedir is the argument to tftpd -s.
#
# To run bootpd by itself, use:
bootps dgram udp wait root /etc/bootpd bootpd
#
# To run dhcpd by itself, use:
#bootps dgram/i udp wait root /etc/dhcpd dhcpd
#
# When running dhcpd and bootpd, bootpd must be run in "slave mode" (with the
# -S option). In this mode, bootpd listens on an alternate port. The port
# bootps-alt is defined to be 950 in /etc/services, but it can be anything
# < 1024. To run dhcpd and bootpd, use the following two lines:
#bootps dgram/i udp wait root /etc/dhcpd dhcpd -b bootps-alt
#bootps-alt dgram udp wait root /etc/bootpd bootpd -S
#
# To run the BOOTP/DHCP relay agent bootpgw, use:
#bootps dgram/i udp wait root /etc/bootpgw bootpgw server-name
#
tcpmux stream tcp nowait root internal
echo stream tcp nowait root internal
discard stream tcp nowait root internal
chargen stream tcp nowait root internal
daytime stream tcp nowait root internal
time stream tcp nowait root internal
echo dgram udp wait root internal
discard dgram udp wait root internal
chargen dgram udp wait root internal
daytime dgram udp wait root internal
time dgram udp wait root internal
pop3 stream tcp nowait root /etc/popper popper
imap stream tcp nowait root /etc/imapd imapd
smtp stream tcp nowait mmdf /usr/mmdf/chans/smtpd smtpd
/usr/mmdf/chans/smtpsrvr smtp
>3. Try "netcat" which bypasses the whole spooler pretzel.
> http://www.cruzio.com/~jeffl/sco/lp/
>
>Sorry about the quick note but I'm a hurry for a free lunch.
>--
>Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
>(831)421-6491 pgr (831)426-1240 fax (831)336-2558 home
>http://www.cruzio.com/~jeffl WB6SSY
>je...@comix.santa-cruz.ca.us je...@cruzio.com
Original E-mail:
Dave Gresham
Jeff Liebermann <je...@comix.santa-cruz.ca.us> wrote:
>On Tue, 6 Apr 1999 19:37:48 GMT, de...@netcom.com (Dave Gresham) wrote:
>
>
>Ok, let me try to decode this cryptic description.
>1. srlaser9 points to port 9101 or port #2 on the EX3 print server.
>2. The error messages occur only sometimes with no obvious pattern.
>3. The error messages ONLY occur with the printer on port #2.
>4. The only email error messages appear from srlaser9. All the other boxes
>and ports apparently work normally and don't complain via email.
>
>Is this correct?
>
>>Maybe it is because of the printer being
>>offline, and in the past year I have upgraded the firmware on the
>>EXPlus 3 in the past year when there was even nastier bugs in it.
>
>If you flashed it within the last year, it's probably current.
>
>If you have SNMP running, invoke:
> getmany printer_ip public iso
>and you will buried in MIB variables and values. The firmware version
>should be on the top line under sysDescr.0. I forgot the exact OID number
>or I would have supplied a "getone" command that only returns the desired
>variable. For instructions on how to make SNMP work on OSR5, see:
> http://www.cruzio.com/~jeffl/sco/snmp_install.txt
>Try it. You might like it.
>
Here is the result of the getmany command.
Name: sysDescr.0
Value: HP ETHERNET MULTI-ENVIRONMENT,ROM D.04.03,JETDIRECT
EX,JD26,EEPROM D.05.20
>>>How do I configure the HP JetDirect EX Plus3 Print Server to use HPNP?
>>> http://www.sco.com/cgi-bin/ssl_reference?105327
>>Already have this.
>
>Good. Now check that the values imbedded in the spooler scripts match
>those in the Tech Article. It's rather odd that only port #2 *SEEMS* to
>having problems. My guess is that either the port is fried on the EX3 in
>question, of there is a typo error in the script where the 9101 port number
>is defined. Watch out for imbeded backspaces and invisible control
>characters. Since you apparently have other EX3 print servers available,
>swapping boxes and reconfiguring the IP's will help point the finger to
>either the box or the Unix configuration.
>
>Remember: Assumption is the mother of all screwups.
>
I checked the Tech article and actually cut and pasted the
lines that needed to be modified and changed the port number
and printer name to be on the safe side. I completely
agree you about assumptions. I let it run for a the last
couple of days, but then i checked this morning and I saw
the same type of email message about this printer.
i did do a diff on HPLaserJet script as someone suggested
and it is the standard file, no differences.
If this client wasn't 2000 miles away I'd go over in
a heartbeat to see if i can swap these things out to
see if there is an actual physical problem with the EX3.
dave
Jeff Liebermann
>Name: sysDescr.0
>Value: HP ETHERNET MULTI-ENVIRONMENT,ROM D.04.03,JETDIRECT
>EX,JD26,EEPROM D.05.20
Gee thanks. Now *I* get to figure out if D.05.20 is current and whether
you have a J2590A or J2593A print server.
Looks like the J2590 software and firmware is missing. The J2593A is at:
http://www.hp.com/cposupport/swindexes/jd2590a_swen.html
and it appears that D.05.20 is current. So much for the firmware theory.
>I let it run for a the last
>couple of days, but then i checked this morning and I saw
>the same type of email message about this printer.
Huh? Days? You only posted one message so I can't look for a pattern.
What are the dates and times the email error messages appear? That should
give a clear clue as to what's happening or who's causing it. If at night
or early morning, that means that someone is probably turning off *ONE* of
the printers or the print server at night. Then, someone else comes along
and tries to print. Methinks you can easily arrange a test to simulate
this condition.
There are also the monster log files in:
/usr/spool/lp/logs/* (or something like that)
which record each print job. Try to get some kind of pattern. Look for
error messages.
>i did do a diff on HPLaserJet script as someone suggested
>and it is the standard file, no differences.
OK. We can eliminate typos. Can I assume that the scripts in:
/usr/spool/lp/admins/lp/interfaces/model.orig/printer_name
are the same for all the other laser printers?
>If this client wasn't 2000 miles away I'd go over in
>a heartbeat to see if i can swap these things out to
>see if there is an actual physical problem with the EX3.
Well, if you're not making any headway, the fact that only one printer on
one print server is screwing up seems to point to a blown port on the print
server. I would pre-program a replacement and mail it to them as the
amount of time we're spending on this is rapidly approaching the cost of a
replacement.
Your previous posting volunteered the absolute minimum useful information.
I'm unwilling to engage in a lengthy Q&A style of troubleshooting via
usenet news. All I can offer is some direction. There has to be a pattern
to the error messages, especially since the exact same printer sometimes
prints properly. Whether the pattern is time, user, cron, application,
configuration, off-line printer, or power on-off related will be shown by
the dates and times of the email messages. It is my contention that the
Unix print spooler is working because the error message indicates that it
has successfully opened the 9101 port for printing. You didn't try my
"netcat" suggestion to bypass the print spooler so this is still a guess.
Closing of the port implies that it was once open. Since the only device
that can normally close the port is the print server, methinks that is the
proper area of investigation. The exact cause of the close is mysterious
but I suspect it will be related to running out of paper, paper jam, going
off line, turning off the power, or some local situation on one of the
three printers or the print server. Methinks simulating all the various
possible problems and looking for email should find the problem.
Good luck.