cancel
Showing results for 
Search instead for 
Did you mean: 

The job is not getting killed

Former Member
0 Kudos

Hi All,

I have one redwood script job which has redwood script in Actions->pre-running.

The job is getting completed successfully in 90% times.But in 10% times it stays running state.

Pre-running state is getting completed means redwood script gets completed and jobs goes to running state.

Here the problem comes and jobs stays in running state and when i try to kill the job the job goes to killing state and it not coming to final state(Completed,error,killed,canceled).

I tries putting the running job to unknown state with help of System_SetJobStatusUnknown but this is not putting the job to unknown state.

Dont know why this job is behaving this way some times.

Any Idea .. Please suggest.

Thanks,

Ravi.

Accepted Solutions (0)

Answers (4)

Answers (4)

Former Member
0 Kudos

Hi All,

Thanks a lot for all your responses.

Once again I would like bring this back to explain this.

This a purely a Redwood job( Mail type job which run in system queue).

As you know in Mail type jobs we write redwood script in Actions->Pre-running or post running.

here I wrote script in Pre-running.

My script in Preruning is getting completed and it generates log file(prerunning.log).

After prerunning the job status changed to running and it stays in same state and it not getting completed or to any final state.

When I tried to end the job using System_SetJobStatusUnkown it is not moving the job to unknown state and generated the log file saying some issue in persisting with respect to time and Timezone.

When I  tried to kill the job it is going to killing state.

My final intention is to remove this killing job or to move it to final state, Later I can debug my code.

Please let me know if you require further details.

Thanks a lot for your efforts!!

Thanks,

Ravi

nanda_kumar21
Active Contributor
0 Kudos

did you try the system_setjobstatusunknown after the job is in killing state?

Last resort would be to set the status at DB level, which i highly recommend not to.

Is there any operator message? If its a mail job, did you receive the mail?

Did you raise an OSS message as recommended?

thanks

Nanda

Former Member
0 Kudos

Hi Ravi,

Please correct me if anything below is wrong. I am also new to CPS

I have also face the same issue.

Even sometimes, with all authorizations in hand, when I try to cancel a job, it will not cancel and will continue to run.

You can stop that job by going to SM66 (Global Work process Overview) and cancel the work process.

This helps in most of the cases.

After cancelling directly in ECC by the above manner, even then in CPS it will be showing as KILLING status.

May be newer versions of CPS may rectify this issue.

Thanks,

Ramakrishnan

h_carpenter
Active Contributor
0 Kudos

Hi Ravi,

If System_SetJobStatusUnknown cannot set a job to unknown, I have never seen this happen, you need to open an OSS message, I am afraid.

Regards,

HP

Former Member
0 Kudos

Carprnter,

Can you tell me what is OSS message?

Former Member
0 Kudos

Please raise message with SAP to resolve this issue.

Former Member
0 Kudos

Hi Ravi,

Please explain clearly as we are unable to get your issue completely.

Regards,

Raja. G

Former Member
0 Kudos

Raja,

The job is running in System Queue and it is in running state and this is not getting completed.

We tried to put the job in unknown state with help of System_SetJobStatusUnknown but no luck the job still stays in running state.

if we kill the job job is going to killing state and not able to set the status to Killed.

Please let me know if u have any answers.

nanda_kumar21
Active Contributor
0 Kudos

Just curious,

Are you running System_SetJobStatusUnknown, when the job is in "running" state or "killing" state? What does the log of System_SetJobStatusUnkown say?

Have you tried running System_SetJobStatusUnkown when the job is in "killing" state? If yes, what does the log say?

Thanks

Nanda

h_carpenter
Active Contributor
0 Kudos

Could you also write down the job id of the job in question and "grep" Default trace for that job id? Also, check if there are any exceptions in the default trace for the time you tried to kill the job/force it to unknown.

Regards,

HP