The Cron daemon terminates

[Rich_Davies]

We had a problem the other day where the cron daemon stopped running any jobs that were scheduled. The last message in the cron log was this:-

cannot allocate 28 byte of space

I restarted the cron and jobs started running again. This has happened once - anybody know why?

[Steven E. Protter · Accepted solution]

Shalom,

Seems like you may be short of disk space or memory.

bdf /var

But if it never happens again, ignore it. If it happens continually, check the cron log and look for a symptom.

SEP

[Bill Hassell]

Also make sure you are up to date on patches for cron and related code.

[spex]

Rich,

You should investigate the last few cron jobs that did run. Something may be amiss which caused a filesystem to fill up.

PCS

[Rich_Davies]

Thanks guys. I think that /var had reached 100%. It's back down to 52% now and working.

[Bill Hassell]

Actually, a lot of processes may have had big problems with /var filling up. Look in /var/adm/syslog/syslog.log for some hints. /var is the most critical directory in HP-UX for availability. /var/tmp is particularly vulnerable since anyone can write to that directory (and must remain that way). I've attached a filesystemk monitoring script that you can put into cron ;-) and it will notify you of almost full status in time to fix it.

[Bill Hassell]

Whoops, here's the attachment.

[Rich_Davies]

Thanks Bill.