Page 1 of 1

Job Reset Issue

Posted: Fri Feb 18, 2011 10:24 am
by DJRH
The job is not getting reset after the abort even though the below option is checked

"Add checkpoints so sequence is restartable on failure".

The error is see in the job is

JobNumber=9, TraceLevel=0/50, JobInstance=1
<<Jobname>> must be reset before it can run


I looked at all the option but does not make sense that why its not doing what it is supposed to do.

Please advise

Posted: Fri Feb 18, 2011 1:02 pm
by chulett
Adding checkpoints gives the Sequence the ability to be restartable, but whether you can or not depends on the Status it ends up in after a failure. It must be in an "Aborted, Restartable" status otherwise you'll manually need to Reset it before you can run it again.

Assuming here we're talking about the Sequence job itself and not any of the individual jobs it may or may not be running.

Posted: Fri Feb 18, 2011 1:15 pm
by DJRH
chulett wrote:Adding checkpoints gives the Sequence the ability to be restartable, but whether you can or not depends on the Status it ends up in after a failure. It must be in an "Aborted, Restartable" status otherwise you'll manually need to Reset it before you can run it again.

Assuming here we're talking about the Sequence job itself and not any of the individual jobs it may or may not be running.
Thanks for your reply,
I understand what you mean.The job is in "aborted/restartable" status and we do not need to reset it to run again the next time when it is supposed to run.

But in my case , the job is in "aborted/restartable" status and when it runs the next day as per the schedule, it does not run and gives the message asking to "Reset" which in case of "Checkpoint" option should not ask to do. The main sequence where the checkpoint is not getting kick off and there are sequence downstream as well but it is not reaching that point.

And yes, we're talking about the Sequence job itself

Posted: Fri Feb 18, 2011 6:32 pm
by greggknight
When A sequence of jobs fails , what is required is that you investigate the Job that failed rerun it and fix the issue THEN you can just execute the sequence and it will start at the last checkpoint.(which is the job that failed and then go on to the remainder of the jobs.

It is not meant for jobs to fail and just keep running the sequence day after hoping someone fixed the issue.

Though I use batch jobs to run all my DS jobs and when a job fails in the batch my batch automatically resets the job but the batch remains failed so that it will not run again until the issue is fixed.

You can set individual jobs to not be checkpointed in a sequence if you wish.
That option is there as well.

Posted: Fri Feb 18, 2011 10:27 pm
by ray.wurlod
Your sequence has a job activity set to "Run" rather than "Reset if required, then run", and the job itself is not in a runnable state for whatever reason (probably an Abort on its previous run).

Posted: Fri Feb 18, 2011 10:31 pm
by chulett
DJRH wrote:The job is in "aborted/restartable" status and we do not need to reset it to run again the next time when it is supposed to run.

But in my case , the job is in "aborted/restartable" status and when it runs the next day as per the schedule, it does not run and gives the message asking to "Reset" which in case of "Checkpoint" option should not ask to do.
I'm afraid that doesn't make sense, so I'm assuming we're missing some vital piece of information here. If your Sequence was indeed in an "Aborted/Restartable" status, then it really should not give the error that you noted when you next tried to run it. What are we missing here? :?

And just to clarify, this 'schedule' you spoke of - is this main Sequence directly scheduled via the Director? In other words, in cron? And you did not resolve the error when it happened but let the next day's run try to pick up where the previous day's run failed? Or was there some kind of attempted intervention between the two runs?