cups scheduler stopping at 4:02 am

[kevin rowe]

found out what's happening at last. the /usr/lib/cups/backends/serial is hanging. Sometimes it won't, if we clear the spool directory, but it ususally does.
We simply disabled it and it is now fine, whatever the cause.

thanks for the suggestion Kurt.

> kevin rowe wrote:
> > I have again looked at this and the logrotate daemon is doing a
> > /etc/init.d/cups condrestart which is hanging after putting out
> > the lines
> >
> > I [28/Nov/2007:04:02:03 +0000] LoadPPDs: Read "/etc/cups/ppds.dat", 2771 PPDs...
> > I [28/Nov/2007:04:02:04 +0000] LoadPPDs: No new or changed PPDs...
> > D [28/Nov/2007:04:02:04 +0000] LoadDevices: Added device "parallel:/dev/lp0"...
> > D [28/Nov/2007:04:02:04 +0000] LoadDevices: Added device "http"...
> > D [28/Nov/2007:04:02:04 +0000] LoadDevices: Added device "ipp"...
> > D [28/Nov/2007:04:02:04 +0000] LoadDevices: Added device "lpd"...
> > D [28/Nov/2007:04:02:04 +0000] LoadDevices: Added device "smb"...
> > D [28/Nov/2007:04:02:04 +0000] LoadDevices: Added device "scsi"...
> > W [28/Nov/2007:04:02:34 +0000] LoadDevices: Backend did not respond within 30 seconds!
> >
> > When I execute the command from the console it will execute
> > fine sometimes, and sometimes it will just hang.
> > There do not seem to be any useful comments about this in
> > Google.
> >
> > Does anybody know which backend is hanging? I'm assuming it's
> > the scsi as it's the last one mentioned.
>
> No, I don't think it is the scsi one. Do you see one that
> is named "snmp" or "dnssd" in your /usr/lib/cups/backend/
> directory?
>
> > Anyone got any suggestions?
>
> Disable all remaining backends (you already know that there's
> no problem with parallel, http, ipp, lpd, smb and scsi) by
> doing a "chmod 000 $backend" [take note of the permissions
> beforehand, so you can restore them].
>
> Test again.
>
> Re-enable each disabled backend one by one, and each time
> test anew.
>
> That way you'll surely find out which is causing that. If it
> is the snmp or the dnssd, you probably don't need them and
> you can disable them permanently.
>
> >> I have a machine that has started (about two weeks ago) to
> > regularly shut down cups in the early morning (04:02 am) and
> > not restart it.
> >>
> >> I cannot see any log activity except /var/log/messages saying
> > cupsd terminating and the cups error log saying stopping for
> > SIGTERM.
> >>
> >> I have scanned all the cron.d directories for jobs that mention
> > cups, and crontab. I've even googled for this.
> >>
> >> I can't find what is stopping cups.
> >>
> >> the cups demon is configured to start at levels 2345
> >>
> >> as far as i know the system has not been changed. From the time
> > i beliewe this is some sort of daily task that is being started
> > because that is when the cron log shows daily jobs starting.
> >>
> >> does anybody have any idea what is doing this?
> >>
> >> thankyou,
> >>
> >> Kevin.
>
>
>
> --
> Kurt Pfeifle
> System & Network Printing Consultant ---- Linux/Unix/Windows/Samba/CUPS
> Infotec Deutschland GmbH  ...........................  Kesselstrasse 19
> A RICOH Company  ...........................  D-70327 Stuttgart/Germany