Subsystem jobs going into error...

Hi ,
I am facing a problem where the subsystem jobs are going into error. AS a result , I have to kill and re-submit the jobs everyday.
Can anybody please help me out with this.....i mean the reason why the jobs are going error...

What do the logs say for the specific subsystem job that is going into error? How about the logs for the queue kernel and subsystem kernel on the box?

What version are you running? I assume you're running 8.9 in a distribution/manufacturing environment and haven't downloaded the 100 or so ESUs to fix memory leaks, crashing business functions, etc.

Please give more details about your environment or we're all just running blind

I am running Xe and not B9

What Service Pack are you running?

[ QUOTE ]
What version are you running? I assume you're running 8.9 in a distribution/manufacturing environment and haven't downloaded the 100 or so ESUs to fix memory leaks, crashing business functions, etc.

Please give more details about your environment or we're all just running blind

[/ QUOTE ]

I am having a quite similar problem. Could you please tell me which ESUs are you refering to?

Diabolical,

I am assuming that you are refering to the Kernel jobs in the OW subsystem, and not the QZDASOINIT jobs in the QUSRWRK subsystem.

I have seen these jobs error out as a result of an outside process or connection ending. TCP services restart. Deployment server goes down for backup (if media objects were being accessed at that time). Security kernel does not relinquish it connections to the F00950 file.

Before you restart the jobs, look at the log and find the time it errored out, see if they all errored out at that time. Then lay down what happens at that time on your architect (backups, scheduled jobs....). If this happens every morning, then chances are it is a scheduled jobs that is effecting them - (like trying to delete SQL packages while services are up). If you want to post a joblog of the errored job - that would hold some answers.