sequence job issues

A forum for discussing DataStage<sup>®</sup> basics. If you're not sure where your question goes, start here.

Moderators: chulett, rschirm, roy

Post Reply
dsscholar
Premium Member
Premium Member
Posts: 195
Joined: Thu Oct 19, 2006 2:45 pm

sequence job issues

Post by dsscholar »

Hi all,

I have a sequence job, in which am running 8 jobs parallely. And i have given an exception handler ---> mail ----> terminator. For testing purpose i made one job to fail. When i ran the sequence job, all the jobs finished except the job with error. How is it possible. I have given a terminator with "abort by sending stop requests " and i dint click "wait for all jobs to finish ".

But all the jobs run for 5 seconds max. I dont know before sending stop requests the jobs might get finished. Please clarify. And when i run the sequence again, its starts should start from the third job(aborted job) but it runs all the jobs. i have enabled "reset if required and run" and reset is happening fine.

1)Reset means previous compiled state?

2)Can anyone tell the significance of "do not checkpoint run". I saw one option like that.

3)How to start the jobs from the point of failure?

4) how to stop running the remaining jobs when one job gets aborted ?



Thanks in advance. your help will be much appreciated.
chulett
Charter Member
Charter Member
Posts: 43085
Joined: Tue Nov 12, 2002 4:34 pm
Location: Denver, CO

Post by chulett »

1) It just means reset, exactly the same as if you had done it manually from Director.

2) When checkpoints are enabled, that option disables them for that particular stage. Meaning it will always run rather than being skipped on a abort/restart even if it ran ok.

3) Enable checkpoints. Make sure your Sequence job aborts when there is a problem such that the status of the Sequence job becomes "Aborted/Restartable". If you run the job again without resetting it first, it will restart from the point of failure.

4. Via the Terminator stage. However, unless something has changed since last I played these games, the Exception Handler only fires when the Sequence itself has a major issue - for example, if it can't find the job to run - but is not triggered by a simple job failure. Those need to be captured by failure triggers and directed to the Terminator, from what I recall.
-craig

"You can never have too many knives" -- Logan Nine Fingers
dsscholar
Premium Member
Premium Member
Posts: 195
Joined: Thu Oct 19, 2006 2:45 pm

Post by dsscholar »

Exception handler makes the sequence job to abort but it allows all others to finish successfully. Thats the problem. But if i give failure triggers for all the jobs and then sequencer(any) -->mail --> terminator it aborts the sequence with abort/restartable when i have given sequence as restartable.

During first run,when one job gets aborted all the remaining jobs doesnt run and tats my requirement. And when i restart it,it starts from the point of failure but eventhough the job fails again, it proceeds with the remaining jobs during the second run. Please advise regarding this.
Last edited by dsscholar on Tue Sep 27, 2011 2:33 am, edited 1 time in total.
dsscholar
Premium Member
Premium Member
Posts: 195
Joined: Thu Oct 19, 2006 2:45 pm

Post by dsscholar »

Does it happen again,as because the jobs are given in paralllel? Should i change it to sequential mode one after another to achieve this logic. Please advise.
dsscholar
Premium Member
Premium Member
Posts: 195
Joined: Thu Oct 19, 2006 2:45 pm

Post by dsscholar »

Hi all,

Please give clarity on "reset if required and run" and "do not add checkpoint" options in sequencer. I have so much of confusions in how its working.

When i just give run in the job activities, the sequence job starts from the point of failure but when i give reset if required and run it starts from the beginning. Please explain this.

In general if you guys give an idea about this it will be useful.
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

Please clarify what you mean by "running 8 jobs parallely". To me it suggests no dependencies between the eight jobs, which means that I'd expect any that did not abort to run to completion (given that they only run for a short time - they finish before the Terminator can send stop signals).
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Post Reply