Restartability of the jobs in Server sequencer

[srinivenigalla]

In a series of hash jobs and sequencers to run for Peoplesoft EPM9.0 in Datastage server edition, I created a Master Sequence in which I am calling all the hash jobs and sequencers using job activities one after another. My questions is to how to restart the jobs in the Master sequencer from the point it failed before. Is there any option avialable and if so, what is it?

'S NEEDED.

[chulett]

Sequence 'General' job properties:

* Enable checkpoints so sequence is restartable on failure
* Automatically handle activities that fail

Those two options are key, especially the first. The second maybe less so since you've already have Sequence jobs with (I assume) error handling built into them. Read all about them in the Online Help and pdf documentation.

You'll need to be on at least a 7.5.x release for it to all work properly.

[Bryceson]

Welcome Aboard :D

Yes, there is two way you can hendle you restartability in sequencer:

1. Add check point on the job level (Job Sequence - Parameter - And check the box Add checkpoints so sequence is restaetable on failure).

2. Add check point on the Project level in Administrator (Sequence Table And check the box Add checkpoints so sequence is restaetable on failure).

Hope that helps!

Bryceson

[Akumar1]

datastage has provided one utility in datastage administrator, in general tab from where you can set some option and you can perform the desired action.and the second thing is you have to check the log and you have to find that under which job sequence a particuler child job has been aborted or encountered warning message after identifying and rectifying the error you have to reset and validate the jobs(need not to compile) encontered with error message or aborted.

Please be careful:
you have to confirm under which job sequenece which particuler job has been encontered with error identify the same job and after resetting start the execution of same job esquence not the master job sequence.


please let me know if you need any further information on the same.


Thanks,
Akumar1

In a series of hash jobs and sequencers to run for Peoplesoft EPM9.0 in Datastage server edition, I created a Master Sequence in which I am calling all the hash jobs and sequencers using job activities one after another. My questions is to how to restart the jobs in the Master sequencer from the point it failed before. Is there any option avialable and if so, what is it?

'S NEEDED.

[chulett]

Technically, only #1 is the 'way'.

2. Add check point on the Project level in Administrator (Sequence Table And check the box Add checkpoints so sequence is restaetable on failure).

This just sets the Project level defaults and ensures that when you create a new Sequence job after it is set that that job will have the option enabled automatically.

[ray.wurlod]

The other way, of course, is to code it yourself in a Job Control routine. This is necessary if you're running a version earlier than 5 (where job sequences were introduced). If you're not, why re-invent the wheel?

[srinivenigalla]

Thanks to all . I tried all the options and still I feel even after all these, the jobs are not triggering if there is an error in the job and not restarting from where the job has been terminated.

[ray.wurlod]

Do you want restart within a job? That is an entirely different question.

Consider the following job sequence.
JobA ----> JobB ----> JobC ----> JobD

Job sequence is executed. JobA finishes successfully and gets a checkpoint. JobB finishes successfully and gets a checkpoint. JobC fails.

When the job sequence is next run, without having been reset or recompiled, JobA is skipped because it has a checkpoint. JobB is skipped because it, too, has a checkpoint. JobC is reset if required, then run. This time it finishes OK so gets a checkpoint. JobD is run and gets a checkpoint. Because all activities finished OK, all checkpoints are removed, so that next run will execute all activities in the job sequence.

Messages are logged indicating which activities are skipped and whether checkpoints were removed at the end.