lpstat consumes a complete CPU in cups 1.4.8
Paul Conklin
paul.conklin at cerner.com
Tue Nov 20 13:49:17 PST 2012
Reliably... no. it's somewhat rare, but we also have over 500+ hosts running CUPS so it usually happens a couple times a month. Very rarely do we touch the daemon. I'm hoping to move to 1.6.x soon (as soon as i can get past my bug using custom filters) so this may be a moot issue if you think its possible it's been fixed upstream.
> Paul,
>
> If you can reproduce reliably by running lpstat at the same time as sending SIGHUP to cupsd, please do file a bug with strace output for the affected PIDs.
>
> On 2012-11-20, at 12:16 AM, Paul Conklin <paul.conklin at cerner.com> wrote:
>
> >>> Weve notice lpstat consuming a full CPU occasionally. Any ideas as to a cause or what we could look at? its on RHEL 5.5 x86_64.
> >>>
> >>> PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
> >>> 15692 d_p0165 25 0 98.9m 2472 1884 R 100.0 0.0 9397:25 lpstat
> >>> 26241 d_p0165 25 0 98.0m 2464 1884 R 100.0 0.0 9408:05 lpstat
> >>>
> >> here are the corresponding pids at the os level
> >>
> >> d_p0165 15692 15667 99 Nov13 ? 6-12:59:43 lpstat -plp62
> >> d_p0165 26241 26181 99 Nov13 ? 6-13:10:23 lpstat -ppahcass1
> >>
> > Looking through my notes for this time, and every other time this is happened, it appears it always seems to co-incide with a cups reload / restart. should I file this as a bug for lpstat? the last time there was a reload was on the 13th. same date as the "zombie" lpstat's"
> >
> > grep reload /var/log/cups/error_log
> >
> > I [13/Nov/2012:10:08:12 -0500] Full reload is required.
> > I [13/Nov/2012:10:08:12 -0500] Full reload complete.
> > I [13/Nov/2012:10:53:22 -0500] Full reload is required.
> > I [13/Nov/2012:10:53:22 -0500] Full reload complete.
> > I [13/Nov/2012:11:04:02 -0500] Full reload is required.
> > I [13/Nov/2012:11:04:02 -0500] Full reload complete.
> >
> >
> >
> > _______________________________________________
> > cups mailing list
> > cups at easysw.com
> > http://lists.easysw.com/mailman/listinfo/cups
>
> ________________________________________________________________________
> Michael Sweet, Senior Printing System Engineer, PWG Chair
>
More information about the cups
mailing list