TM1 Process Keeps Rolling Back and Restarting in Parallel Executions of Batch Mode

Post Reply
User avatar
mce
Community Contributor
Posts: 300
Joined: Tue Jul 20, 2010 5:01 pm
OLAP Product: Cognos TM1
Version: Planning Analytics Local 2.0.x
Excel Version: 2013 2016
Location: Istanbul, Turkey

TM1 Process Keeps Rolling Back and Restarting in Parallel Executions of Batch Mode

Post by mce » Mon Feb 04, 2019 2:09 pm

Hi,

We use IBM Planning Analytics Local v.2.0.5.

We have a lot of data that we need to pull into TM1 cube from Oracle DB.
Hence we use batch mode with commands BatchUpdateFinishWait(0) and BatchUpdateStart to load them in parallel threads to make it loaded in shorter time.

A parent process triggers the same data load process many times with different parameters for them to be run in parallel via TM1RunTI, and the parent process also waits for the last one to finish.

Each parallel execution of the process is updating a different slice of the same cube.

For some reason, sometimes some of the parallel batches rolls back and starts again.
We could not understand why it rolls back and restarts. This issue increases the whole execution time.

What might be causing this issue? Any ideas?
Thanks in advance for all comments.

Regards

mvaspal
Community Contributor
Posts: 328
Joined: Wed Nov 03, 2010 9:16 pm
OLAP Product: tm1
Version: 10 2 2 - 2.0.5
Excel Version: From 2007 to 2013
Location: Earth

Re: TM1 Process Keeps Rolling Back and Restarting in Parallel Executions of Batch Mode

Post by mvaspal » Tue Feb 05, 2019 8:54 pm

I saw this in 2.0.6 when the process clashed with a SavaDataAll chore and in that case either both were hanging forever or in better case the process with batch update rolled back and ran once again

User avatar
mce
Community Contributor
Posts: 300
Joined: Tue Jul 20, 2010 5:01 pm
OLAP Product: Cognos TM1
Version: Planning Analytics Local 2.0.x
Excel Version: 2013 2016
Location: Istanbul, Turkey

Re: TM1 Process Keeps Rolling Back and Restarting in Parallel Executions of Batch Mode

Post by mce » Sat Jun 15, 2019 8:42 am

I understand that this issue might be due to locks on CAM user logins, as we use a CAM user for executing parallel threads via RunTI. I will check this in PA 2.0.7 with RunProcess or with OptimizeClient parameter and will see if the same problem happens there.

Post Reply