Problem with OWB Mapping - hanging in there !

[uparna]

Folks, i am back with a problem.

We have an OWB mapping that is configured to do ETL operations on a large amount of data. When we tested this in our development environment ( which had around 50000 rows), it ran just fine - in production though, since we have around 500000-600000 rows(10 times devl), the mapping is going bonkers.

It simply "hangs in there".

I noticed that all the ETL tasks are completed - ie data is loaded in its entirety - but the process is simply not marked as "completed" - so it kinds hangs in there.

i tried reducing the data size by 1/6th , and the process ran peachy-fine.

dont aha! yet....

i retained the original workload and simply configured the OWB mapping parameters to not put in any process metadata and - with the same original workload - the process completed successfully.

hallelujah.

Now we can surmise that this has something to do with both the volume and the metadata ( me = genius) . Has anyone faced any such situation before with such OWB "bad" behaviors?

Database Version = 10.2.0.2.0

( PS - have a TAR for this - but Oracle asked us to change 2 parameters - aq_tm_processes (from 0 to 1) and job_queue_processes (from 6 to 10) -> these changes have failed to solve the problem ).

I know you need a ton of information before you can help - all i can say is "aye aye captain , i am ready !"

Regards,
S. Jayaram Uparna .

 

[uparna]

bump

Regards,
S. Jayaram Uparna .

 

[uparna]

Regards,
S. Jayaram Uparna .

 

[johnherman]

I have not seen much traffic in this forum for OWB. You might want to consider posting in the Oracle 10g forum.
forum1177
Good luck

-------------------------
The trouble with doing something right the first time is that nobody appreciates how difficult it was - Steven Wright

 

[uparna]

had put it there before this thread itself - no responses yet -- and the clock is ticking.... sigh :-(


thanks anyway !

Regards,
S. Jayaram Uparna .