DSXchange

Hi,

I am trying to use Exception Handler along with Terminator stage.

There two jobs which run parallely, to fecth data from two different tables.
Now, the idea is to abort the master sequence immediately, after either of the jobs has aborted.

When I am using Exception handler along with Terminator, I am acheiving what I want, but what is confusing is that the job status of the other job ( the job for whcih terminator stage send sthe request ) is sometimes abrting with Aborted status and sometimes with Stopped status.

I am selecting, Send STOP request to all running jobs.
I also tried to select And wait for all jobs to finish.


Please explain.

Regards,
Asit

Hi,

The Sequence design is this job design.

The idea behind using the Exception handler is to stope execution of all running jobs in case of any Job / Sequences in this sequence abort.

Now, in my case this is what is happening:

The Job ExtractEPMData aborts and this is caught by Exception handler.
the exception handler triggers the Terminator stage, which sends STOP requests to all running jobs and waits for them to finish.

Now, this Seq_FetchGl sequence upon receiving a STOP request sometimes, finishes is Stopped/restartable and sometimes in Aborted/restartable state.

Upon restart of the whole sequence, so if the Seq_FetchGL is in Stopped/restartable state, it does not reset and restart ( I have set the Execution action as Reset if required, then run ). and hence the controller timesout with the error message Error Timeout waiting for event, ErrCode = -14.

But had the Seq_FetchGl finished in Aborted/restartable state, then it restarts smoothly and no problems to me too...


I dont understand this behavior.

Please explain and guide me.

Thx.

A Stop request is just that, a request, to be processed by the job next time it pays attention to its RT_CONFIG table. There is no such thing as an immediate stop.

The job may indeed finish before it gets around (would have gotten around) to handling the stop request.

And yes, sometimes a Stop request makes a job stop and sometimes it makes it abort. You'd need to ask IBM why it works like that. And speaking of 'why' - from what I've seen, 'Stopped/Restartable' is not in fact restartable. Only an 'Aborted/Restartable' Sequence job can actually be restarted properly from the last checkpoint. Something else to ask about...

Do u guys agree that Stopped/Restartable is not a restartable state ?

If such be the casem then can u plz guide me for handling this requirement :

" If two jobs start parallely, and either of them aborts, then the other must also abort, such that both the jobs are restartable."

Reject the requirement. It can not be reliably achieved.

Learn English. The second person personal pronoun is "you", not "u". Plz, I believe, is a city in Borogravia.

Sorry for the wrong English usage.

I would try for handling this requirement in some other way... shall get back once some solution is there...