#1050 cronjobs hanging
Closed: Fixed None Opened 15 years ago by ricky.

Some cronjobs seem to be hanging on a bunch of our machines, which is blocking certain important jobs as well. For example, on serverbeach5 right now:
{{{
[root@serverbeach5 ~]# ps -ef | grep cron
root 1203 2504 0 Dec10 ? 00:00:00 crond
root 1204 1203 0 Dec10 ? 00:00:00 /bin/bash /usr/bin/run-parts /etc/cron.daily
root 1303 1204 0 Dec10 ? 00:00:00 /bin/sh /etc/cron.daily/rpm
root 1304 1204 0 Dec10 ? 00:00:00 awk -v progname=/etc/cron.daily/rpm progname {????? print progname ":\n"????? progname="";???? }???? { print; }
root 2504 1 0 Dec05 ? 00:00:00 crond
root 6082 2504 0 Dec08 ? 00:00:00 crond
root 6083 6082 0 Dec08 ? 00:00:00 /bin/bash /usr/bin/run-parts /etc/cron.daily
root 6088 6083 0 Dec08 ? 00:00:00 awk -v progname=/etc/cron.daily/logrotate progname {????? print progname ":\n"????? progname="";???? }???? { print; }
root 9645 2504 0 08:24 ? 00:00:00 crond
root 9858 9821 0 08:37 pts/0 00:00:00 grep cron
root 19654 2504 0 Dec09 ? 00:00:00 crond
root 19655 19654 0 Dec09 ? 00:00:00 /bin/bash /usr/bin/run-parts /etc/cron.daily
root 19754 19655 0 Dec09 ? 00:00:00 /bin/sh /etc/cron.daily/rpm
root 19755 19655 0 Dec09 ? 00:00:00 awk -v progname=/etc/cron.daily/rpm progname {????? print progname ":\n"????? progname="";???? }???? { print; }
root 26455 2504 0 Dec07 ? 00:00:00 crond
root 26456 26455 0 Dec07 ? 00:00:00 /bin/bash /usr/bin/run-parts /etc/cron.daily
root 26562 26456 0 Dec07 ? 00:00:00 /bin/sh /etc/cron.daily/rpm
root 26565 26456 0 Dec07 ? 00:00:00 awk -v progname=/etc/cron.daily/rpm progname {????? print progname ":\n"????? progname="";???? }???? { print; }
}}}

This has already caused annoyance by locking the rpmdb on a machine twice.


This is actually not a hanging cron issue but a rpmdb issue. Both sb4 and sb5 have issues keeping local database copies in proper order. We've seen corruption in xenstored files and rpmdb files. It is quite strange.

Seth did some work on these boxes that seems to have fixed this issue. I'm going to close it. We can re-open if it comes back. Which I think it likely will.

Login to comment on this ticket.

Metadata