User Crontabs Not Running...

Hello everyone,

I am running openSuSE v10.2, cron-4.1-68, and an application that I use creates a user crontab file (/var/spool/cron/tabs). The problem is that the crontab is not running anymore.

I added a date command piped to a file every 30 min. to see if cron is running my scripts, which it is not.

I prod cron by both 'crontab -u [username] -e' adding a blank line to the top forcing a save and a 'rccron restart'. It may run for a while but then give up, or not run at all.

When I add entries into the /etc/crontab file directly, there are no issues at all and everything runs smooth... but then I am constantly making manual changes to etc/crontab when my application changes its user crontab.

There must be a reason for this... I can't find anything in the logs either that suggest a syntax issue or such. Any suggestions??

Thanks ever so much!
-Cheers, Peter.

Remember that the syntax of user crontabs differs from root crontabs.

/etc/crontab and crontabs in /etc/cron.d look like this:

0 0 * * * root command arg1 arg2 ...

User crontabs created by crontab -e look like this:

0 0 * * * command arg1 arg2

Basically users have no option to run as anybody except themselves, whereas root crontabs can switch users.

ken_yap,

Thanks for the response...

Yes, that is the format that I am using. Here is my date command for checking:
* * * * * date >/data/zimbra_cront.lst

For instance, my date file shows that the crons last ran yesterday around 11:34pm. I did an rccron restart this morning and its running again, but I don't know when it will stop.

Thanks.

-Cheers, Peter.

Before you restart the cron service, check if /usr/sbin/cron is running. If it is, do a strace on the process id to see what it's doing, why it's stuck.

Will do...

Right now its not hanging (seeing as I am watching it like a hawk)... but I will try and catch it in the act (or lack of act)...

Thanks. -Cheers, Peter.

Well, it died again last night... my file contains 'Fri May 23 00:14:01 EDT 2008'

The process was dead. Again, nothing noticeable in the logs... arg.
I will look to see what runs at that time.

-Cheers, Peter.

Attach a strace to the running cron to see what why it dies.

Omg... :blink:
ken_yap... Thank you ever so much and so sorry for wasting your time. I feel like hitting myself with a stupid bottle. It was my (very stupid) bad.

I was digging in the wrong hole. I ran strace like you mentioned, several times I might add, because I could not figure why cron was getting a kill signal. Well, my stupidity... to work around an application bug I was restarting syslog-ng in another crontab... or so I thought. For the syslog restart line I had entered cron... so I basically had a line that killed cron. I just am in awe of this...

Me thinks a vacation is in order... ohi.

Thanks ever again for your help.
-Cheers, Peter.

Don't worry about it. We all do this sort of thing sometime. It's human.