cron jobs die after 5-6 hrs

[dominic_7]

hi ,

we have a Tru64 cluster with Alphaservers ES80 and GS60 , running Tru64 unix 5.1B , pk3 .
Since last one week cron jobs have been getting killed suddenly after 5-6 hrs .
The cron daemon continues to run unaffected.
however the cron jobs terminate .
all permissions are unchanged .
This is the first time in 3 yrs we're getting this kind of a problem .
Apparently no changes were made to the system in the last few weeks .

can anyone advise please

regards

Dominic

[Venkatesh BL]

Does the syslogs or binary error log files reveal anything?

[Michael Schulte zur Sur]

Hi,

the description is a little unspecific. The job run and then disappear, right?
Where is the output of the jobs going to?
Is there evidence that the jobs get killed by the kill command or by any lack of resources?

greetings,

Michael

[dominic_7]

hi ,

No theres no evidence that the jobs get killed by the kill command .
Nor is there any resource crunch on the system .
An there are no error messages in the
messages file .
I have to check the binary file and syslog .

Could this ber related to max jobs parameter in the /var/adm/cron/queuedefs
file ? where in maxjobs=25
and here users like oracle keep adding cronjobs ?

thanks

dominic

[Michael Schulte zur Sur]

Have you checked the cron log file?
If 25 job isn't enough it should show up there.
Also in that case job are not started at all.

Michael

[Ivan Ferreira]

You should check the cron log file (/var/adm/cron/log), and also, you should configure your cron job to redirect stdout and stderr to a log file.

You can check the mail of the user that runs the cron to verify if there was some output.