Friday 19 July 2013

Common Process failures

The App Engine program has run to No Cuccess and we dont see any error in the log.
But as per the trace file that is being generated we can see that the following errors have occurred:


28 01.00.04 340.792628 CMgrUpdateTransaction::OnPreCommit : trying to commit a failed transaction
29 01.00.04 0.000823 CMgrUpdateTransaction::OnPreCommit : trying to commit a failed transaction
30 01.00.04 0.008131 CMgrUpdateTransaction::OnAbort
31 01.00.04 0.000033 CMgrUpdateTransaction::OnAbort
32 01.00.04 0.000738 CMgrUpdateTransaction::OnPreCommit : trying to commit a failed transaction

Cause:
The cause of this issue is corruption of the Process Scheduler cache on DB Platform.
This is sporadic failure, some App Engine programs will fail and some of them will run successfully.

Steps to be taken:
Try restarting/ rerunning the processes after deleting the failed instances.
If the problem persists, Bounce the Process scheduler server and Clear Cache.


SOLUTION INSTRUCTIONS
---------------
Implement the following instruction steps:
1. On Process Scheduler Windows box, run psadmin to stop the Process Scheduler.
2. Navigate to PS_HOME\appserv\prcs\DBNAME directory.
3. Delete the CACHE directory.
4. Retest the issue.
5. Migrate the solution as appropriate to other environments.

No comments:

Post a Comment