It's all about the answers!

Ask a question

why if using RI I must disable the data collection jobs (ETL) of the CLM applications?


Silvia Mesturino (1162129) | asked Oct 01 '13, 8:00 a.m.
I found this info in the IBM Help:
If you upgrade to Rational® Insight 1.0.1.1 and use the Data Manager data collection feature of Rational Insight, you must disable the data collection jobs of the CLM applications.
My system is organized in 6 databases: I_CCM, I_ETL, I_DW, I_RRC, I_QM, I_JTS: which databases are updated by the DM ETL jobs and which ones are updated by CLM ETL jobs ? which database is used by the RQM Built-in BIRT reports ?

Accepted answer


permanent link
Abraham Sweiss (2.4k1031) | answered Oct 01 '13, 8:10 a.m.
All the etls that feed into the data warehouse that is being upgraded need to be disabled during the upgrade process.  This is to ensure that no etls are running while the data warehouse DB is being upgraded.  In short all etls feed the data warehouse DB, so they all need to be disabled.

Once the upgrade is completed, then re-enable the etls.
Silvia Mesturino selected this answer as the correct answer

Comments
Silvia Mesturino commented Oct 01 '13, 8:17 a.m.

Thank you, I had misunderstood the message, since a collegue of mine says that I need to shut down all CLM ETL jobs, now that he has enabled the RI DM ETLs.
Can you also answer the part of the question regarding which data base is updated by which ETL job ?

4 other answers



permanent link
Abraham Sweiss (2.4k1031) | answered Oct 01 '13, 8:27 a.m.
BOth etls:
1) Feed into the same data warehouse DB
2) Basicly do the same thing
) Running both could cause conflicts

The advantage of running the Insight ets is that they run on the insight jvm, isolated from the CLM jvms.  If the CLM etls are run, they will run on the same jvms as CLm and use up resources that could otherwise be used for other CLM activities.





Comments
Silvia Mesturino commented Oct 01 '13, 8:34 a.m.

OK, that could be the reason, if both were scheduled to run at the same time.
Now, can you tell me why (I have both sets of ETL jobs running) and still I do not have the right data in the RQM report (trend execution) and I need to run the RQM ETLs manually in addition almost every day since a week ? Is all data in I_ETL or or I_DW or I_QM ?


permanent link
Abraham Sweiss (2.4k1031) | answered Oct 01 '13, 8:41 a.m.
i do not believe it matters when the insight etls or CLM etls run.  If they both run, there is potential for conflicts.  

If the  star jobs are completing successfully and the data appears to be missing information, then please open a PMR to engage support.


Comments
Silvia Mesturino commented Oct 01 '13, 8:46 a.m.

This is also a question I wanted to ask. Is the star job the one that matters ? The star job is configured to run under Insights and is successfull. Thank you !


permanent link
Jackie Albert (1.6k14744) | answered Oct 01 '13, 9:10 a.m.
You actually need the main RQM job to successfully complete as well as the Star job.  The main RQM job will populate the operational data store (ODS) section of the data warehouse with a snapshot of the most recent data.   Once that has completed, the Star job uses the ODS data to calculate its metrics, so if you don't have both jobs completing successfully some of your metrics may be incorrect, such as your execution trends.

Comments
Silvia Mesturino commented Oct 01 '13, 9:23 a.m.

Does the sequence matter ? I see the QM job (QM Job for my PA) run short after midnight, while I see the star job, configured in RI run at 2 am. In the morning I then have to run teh QM job again otherwise the old snapshot does not contain all data.


Jackie Albert commented Oct 01 '13, 9:28 a.m.

Yes - the QM job should run first. Otherwise the ODS will not be populated correctly for the Star job to gather the metrics.


permanent link
Abraham Sweiss (2.4k1031) | answered Oct 01 '13, 8:48 a.m.
For RQM, my understanding is that the STAR job is the one of interest.

Your answer


Register or to post your answer.