jobs run in one run and abort in an other run(without change

Post questions here relative to DataStage Server Edition for such areas as Server job design, DS Basic, Routines, Job Sequences, etc.

Moderators: chulett, rschirm, roy

Post Reply
Pavan_Yelugula
Premium Member
Premium Member
Posts: 133
Joined: Tue Nov 23, 2004 11:24 pm
Location: India

jobs run in one run and abort in an other run(without change

Post by Pavan_Yelugula »

hi all
we have perfectly running jobs in one run getting aborted in a other run
This message is being shown in the log:


Parallel job reports failure (code Internal error - missing script file RT_SC148/OshExecuter.sh)

Is there something i can ensure to make it a fool proof run

Any inputs will be really helpful

Thanks and Regards
Pavan
ArndW
Participant
Posts: 16318
Joined: Tue Nov 16, 2004 9:08 am
Location: Germany
Contact:

Post by ArndW »

Pavan,

please give the forum some more information. You have told us that sometimes the jobs run and sometimes they don't . The .osh are orchestrate shell scripts which are generated at compile-time so each invocation should find the same file - unless something else is happening.

Can you explain a bit more in detail? Do the jobs always run the first time? Have you looked at the log files in detail when the job aborts? What does the job do? What sort of HW-env do you have? Etc..
Pavan_Yelugula
Premium Member
Premium Member
Posts: 133
Joined: Tue Nov 23, 2004 11:24 pm
Location: India

Post by Pavan_Yelugula »

Sorry
For being ambigous on this.

The jobs which are running in one-run doesn't run in the other run.
They run the first time and when i try to run them again it aborts giving the RT Error.
If i go in and again manually compile all such kind of jobs which are throwing this error then they again start running.

Then they run for a couple of runs and the problem crops up again.

These are pretty straight forward jobs which just read some data from the database and load it in to files and i also have jobs which just do some lookup and some transformations and even they fail intermettitenly without even touching these jobs.

we use Datastage 7.1.1 and the jobs which fail are parallel jobs and we have Informix Database which holds pretty good amount of data.

Any inputs on this will give a hell of a credibility to us in front of the client because no one going to accept a product which even the developer doesn't know when it is going to run and when it is not going to run

Thanks and Regards
Pavan
Pavan_Yelugula
Premium Member
Premium Member
Posts: 133
Joined: Tue Nov 23, 2004 11:24 pm
Location: India

Post by Pavan_Yelugula »

Hi all
Can atleast any one tell me where to search the answer for this one
dsxuserrio
Participant
Posts: 82
Joined: Thu Dec 02, 2004 10:27 pm
Location: INDIA

Post by dsxuserrio »

Pavan
Try to copy the existing job as a newjob.
Test to Test1.

Delete Test.
Rename Test1 back to Test. Then compile and run.

Hope this helps.

Thanks
dsxuserrio

Kannan.N
Bangalore,INDIA
Post Reply