From: Shaun.Racine@intier.com
Date: Tue Oct 14 2003 - 04:21:56 EDT
Sort of summary - sorry for delay.
I got one response, from Mark Sornson, much appreciated. He suggested
applying PK 5, which fixes some printing issues.
Original question and Mark's interspersed replies below.
Our local IT at site will apply a patch over the Christmas period.
It was the first time we had this problem, and it has not recurred (yet).
If it does occur again we will have to decide whether to switch on
lpr.debug :(
Regards,
Shaun Racine
----------------
Mark Sornson:
Someone (from another group at work) forwarded your
query to me.
> To: tru64-unix-managers@ornl.gov
> CC:
> Subj: Unix Printing - No lpd daemon running
>
[snips on header info]
> Hi Managers,
>
> Tru64 v5.1A, patch kit 3 (T64V51AB03AS0003-20020827). ES40 / HSG80 LSM
&
> AdvFS.
Patch Kit 5 exists that has some printing fixes you might want
to pick up, unless your site is very sensitive to upgrades.
>
> One of our sites had a problem with printing after a reboot. The server
is
> scheduled to reboot at the same time each week, with no previous problems
> in this area.
Is it correct to assume that this is the first time you've
seen this problem (and that it hasn't returned)?
>
> All printing in a queued state.
> Tried to restart specific printer queues but no joy.
>
> lpc> restart r5_cp_2
> r5_cp_2:
> no daemon to abort
> r5_cp_2:
> lpc: connect: No such file or directory
> couldn't start daemon
> lpc> quit
>
> No lpd daemon running.
> Ended up starting the lpd deamon manually.
>
> root> /usr/sbin/rc3.d/S65lpd start
>
> All printing sprung into life again and remained stable.
> No errors in startup logs, typical message (and "NOTICE" when manually
> started) from lpr.log;
> Sep 14 19:48:34 rimmer lpd[59924]: ERROR -- startdaemon: connect failed
to
> socket /dev/printer, printer r5_cp_2, error: No suc
> h file or directory
> Sep 14 19:51:43 rimmer lpd[60121]: NOTICE -- lpd: MASTER DAEMON STARTED
Hm ... are the above entries all that were in lpr.log?
That clearly suggests that the parent daemon either
didn't start at reboot or crashed early on.
I hate to suggest it, but setting lpr logging to
"lpr.debug" in /etc/syslog.conf may reveal more
information. But you shouldn't bother to do it
until just before you reboot, and you should probably
set it back to its original value after you reboot
(after printing starts), so you don't eat up disk
space with unnecessary logging messages.
>
> Does anyone know what may have caused this? Or where to look for more
> information?
The lpr.log file with "lpr.debug" set is your
best bet for additional information. Or look
in / for a core file (from lpd), or in any of
your printing spool directories for core files.
You say you started lpd by hand (/sbin/init.d/lpd start).
When no jobs are running, you don't have a problem
getting printing to restart if you bounce it by hand,
do you (/sbin/init.d/lpd stop and then start)?
-mark.
This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:49:38 EDT