Hello.

We've been having in issue on a few printers where a large print job
will stop printing, and then the printer becomes unusable for a
considerable amount of time (30+ minutes). We've been able to reproduce
the error by sending a large job to the printer and then pulling the
power... initiating a paper jam.. The result is a job in the queue that
holds up all the incomming print jobs. Cancelling this job will allow
the print jobs to come through.

Is there a reason that these jobs are not automatically restarted once
the printer comes back on line? More importantly, is there a way to
resolve the issue other than manually cancelling the stuck jobs?

Thanks for any help you can provide.

Luke MacNeil -

[ Post a follow-up to this message ]

On 3 Nov 2005 08:20:29 -0800, Luke MacNeil wrote:
>
> We've been having in issue on a few printers where a large print job
> will stop printing, and then the printer becomes unusable for a
> considerable amount of time (30+ minutes). We've been able to reproduce
> the error by sending a large job to the printer and then pulling the
> power... initiating a paper jam.. The result is a job in the queue that
> holds up all the incomming print jobs. Cancelling this job will allow
> the print jobs to come through.
>

You might get an answer if you provide the following informations:
- What kind of printer (maker/modell)
- How is this printer connected to your print spooler
(serial/parallel/network)
- Which UNIX(like) operating system are you using
- What printspooler are you using (CUPS/LPRng/whatever comes with your
OS/...)
- What is "large", what is it and how does "large" relate to the printer
memory.

Other than that I would suggest you try to figure out the root cause - the
printer stopping amidst a job. This should not happen in the first place.

Helmut

--
Almost everything in life is easier to get into than out of.
(Agnes' Law)

[ Post a follow-up to this message ]

Its an HP4050
The printer is connected via a Jetdirect box... (ethernet)
Running Solaris 8.
I'm not sure what the printspooler is... how do I tell? I'm using the
solaris lp/lpstat/cancel commands to work with the queue.
Large meaning enough pages for me to start the job and walk over to
pull the plug...
The root cause is that the printer stops printing.. may it be paper
jam, out of paper, power failure....Wrong paper size...

My issue is that the spooler does not restart printing the job, and it
holds up the queue until it is manually cancelled.

Thank you for your response.

Luke.

[ Post a follow-up to this message ]

Luke MacNeil wrote:
>
> Its an HP4050
> The printer is connected via a Jetdirect box... (ethernet)
> Running Solaris 8.
> I'm not sure what the printspooler is... how do I tell? I'm using the
> solaris lp/lpstat/cancel commands to work with the queue.

That does answer the question.  If you don't know what other
spooler you're using then given those commands it is the
out-of-box System 5 spooler that comes by default on Solaris.

Look in /var/lp, /var/spool/lp, /usr/lib/lp for assorted scripts,
binaries, config files, pipes, queuing directories and so on.
One interesting thing to find is the directory under /var/spool/lp
that stores the binary being printed and the current page
offset.  The small file that stores the page offset does allow
for partial print jobs on restart given reading a ton of man
pages.

> Large meaning enough pages for me to start the job and walk over to
> pull the plug...

On that size printer we're talking 20+ pages.

> The root cause is that the printer stops printing.. may it be paper
> jam, out of paper, power failure....Wrong paper size...

You're confusing root cause with effect.  The printer stops
printing, that's the effect.  Paper jam, out of paper, power
failure, that's the root cause.  Which one triggers the problem
or do all of them?

> My issue is that the spooler does not restart printing the job, and it
> holds up the queue until it is manually cancelled.
>
> Thank you for your response.
>
> Luke.

[ Post a follow-up to this message ]

>You're confusing root cause with effect.  The printer stops
>printing, that's the effect.  Paper jam, out of paper, power
>failure, that's the root cause.  Which one triggers the problem
>or do all of them?

All of them trigger the current job to stay in the queue until its
manually cancelled.

>Look in /var/lp, /var/spool/lp, /usr/lib/lp for assorted scripts,
>binaries, config files, pipes, queuing directories and so on.
>One interesting thing to find is the directory under /var/spool/lp
>that stores the binary being printed and the current page
>offset.  The small file that stores the page offset does allow
>for partial print jobs on restart given reading a ton of man
>pages.

Thank you, I'm going to spend some time on this today, I'll let you
know if I find anything.

Thanks again for your help,

Luke.

[ Post a follow-up to this message ]

I recieved this message today:

The printer MFG1_9000 has stopped printing for the reason given below.
Fix the problem and bring the printer back on line.
Printing has stopped, but will be restarted in a few minutes; issue an
enable command if you want to restart sooner.
Unless someone issues a change request

lp -i MFG1_9000-4762 -P ...

to change the page list to print, the current request will be reprinted
from the beginning.

The reason(s) it stopped (multiple reasons indicate repeated attempts):

exec exit fault


I have never seen this message before. Although the print job started
again on its own, I dont know how long it took to do so..Is there a log
file that would hold that information somewhere?

[ Post a follow-up to this message ]

"lpsched" 6 lines, 286 characters
11/08 07:06:26: printer fault. type: mail root, status: c
msg: (hpnpf (read bytes from socket): Connection reset by peer)
11/08 07:06:31: printer fault. type: mail root, status: 14
msg: (exec exit fault)
11/08 07:07:35: printer fault. type: mail root, status: 14
msg: (exec exit fault)


This is the /var/lp/lpsched log.. but it doesn't say when printing
resumed.. the same error twice in one minute..

[ Post a follow-up to this message ]