SUMMARY: 2.6 cron not executing jobs, logs "queue max run limit reached"

From: sun-managers-redist@treehouse.napa.ca.us
Date: Wed Sep 02 1998 - 10:34:21 CDT


My original question:

> I just set up some new U5s with 2.6. On one machine, /var/cron/log
> was over 4MB and had messages like:
>
> ! c queue max run limit reached Tue Sep 1 11:42:00 1998
> ! rescheduling a cron job Tue Sep 1 11:42:00 1998
>
> The jobs listed in root's crontab are not getting executed any more.
>
> I ran /etc/cron.d/logchecker, which renamed /var/cron/log to
> /var/cron/olog, but cron continued writing to /var/cron/olog.
>
> Can anyone tell me how to cause cron to start executing jobs again?

The problem turned out to be a poorly-behaved job that didn't
exit (and was getting called several times an hour), thus
causing lots of children cron was awaiting. Corrected the job
script to exit properly, killed off the muiltitudinous jobs,
and things returned to normal.

Thanks, everyone!

 ~!paul

------------------------------------------------------------------------

From: Craig Raskin <raskin@compusec.org>

You need to kill -1 the daemon. If that doesn't work, just kill and
restart it.

------------------------------------------------------------------------

From: Stephen Harris <sweh@mpn.com>

You have old cron jobs running that haven't terminated. cron can only
run a certain number of jobs per queue, so if your cronjobs never terminate,
cron will attempt to reschedule pending jobs.

Find the broken non-terminating program and kill it :-)

------------------------------------------------------------------------
From: anders@hmi.de (Thomas Anders)

/etc/init.d/cron stop; cp /dev/null /var/cron/log; /etc/init.d/cron start

What patch level is your cron? Beware that 105393-05 is said to be broken.
105393-06 should be the latest and greatest.

------------------------------------------------------------------------
From: Casper Dik <casper@holland.Sun.COM>

run ptree on cron and check its children, apparently a number of
jobs hasn't terminated yet.

See queuedefs(4) if you need to increase the number of cron jobs,
but it's most likely a bug in one of your jobs.

------------------------------------------------------------------------
From: Ray Trzaska <rtrzaska@uk.mdis.com>

it sounds like you have more than 100 jobs trying to run
simultaneously from cron!

cron jobs run in the batch queue 'c'

check out that you haven't redefined the attributes of that queue
see /etc/cron.d/queuedefs
and check out the man page for queuedefs

------------------------------------------------------------------------

From: jbwendl@bnpcn.com (James Wendling)

I beleive you have reached the limit fo concurrently executing cron jobs. When
the amount of running jobs decreases new ones will be allowed to run.

------------------------------------------------------------------------

From: Leif Ericksen <phantom@wwa.com>

Try /etc/init.d/cron stop then /etc/init.d/cron start

That will start and stop cron without you having to kill it. The cron
should program should be able to start working again with the crontab
<user> construct as well...

------------------------------------------------------------------------

From: "P Wallis" <p.wallis@x400.icl.co.uk>

Had similar problem on E450 & 2.6.
If you cannot reboot system kill the cron and manually
restarted.

--
G. Paul Ziemba  paul@w6yx.stanford.edu

4.3 BSD unix: 8:31am up 19 days, 17:07, 1 user, load average: 0.23, 0.18, 0.02



This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:12:48 CDT