DSXchange

Hi Guys,

We have an ISD enabled parallel job which is calling a Sequencer with the help of Job Controller, now the problem is whenever we are trying to compile the Sequencer, it says, it is being monitored even after undeploying ISD service.

Job flow is something like below --
ISD Service --> Parallel job (Job Crontroller)--> Sequencer Job

Here, Parallel job is only having Job controller code which is triggering the Sequencer. Admin team is unable to trace the locks and restarted the DS services to unlock the job but this is not a feasible solution in PROD

Kindly suggest if someone faced this issue before.

Did the admin team try to clear the locks with the cleanup_abandoned_locks.sh utility from /opt/IBM/InformationServer/ASBServer/bin? This might help specially if they can not see an active session in the webconsole or via the traditional Unix commands

No, in version 11.5 this script is not available.

...try below from same location

./xmetaAdmin.sh cleanupAbandonedLocks -dbfile ../conf/database.properties

Notice that you could also list and kill all user and trusted active sessions, including the sessions that are not visible from the web console, using the SessioAdmin.sh utility from same directory, but you are trying to clean up locks in production environment .. just need to be sure and extra careful

Thanks Julio, I'll ask my admin team to execute this script and update with the outcome.

But any idea why my sequncer is getting locked? This sequencer is not being called by the service as design is:
service --> parallel job(job controller) --> this sequencer

The most common cause for DataStage jobs to get locked is leaving them open or force them to terminate, but your issue is in production environment where you wouldn't need to compile your jobs unless is your standard practice

What's the topology of your ISD job? I would guess that it executes the DS Sequence one to completion...right? Is the job Sequence defined as multi instance? Give us more details to be able to figure out the root cause of your locking issue

Regards

JRodriguez,

My service is invoking a parallel job which has a job controller. This job controller code is calling a sequencer which calls other job
After successful completion of sequencer , it will generate a status file, which has been returned by the service on display to invoker.

Service --> Parallel job (Job Controller) ---> Sequencer ---> Generate Status file ---> Read by Parallel job (Job Controller) ---> return the status to Invoker