DB2 V7 Z/oS : Archive Log MODE(QUIESCE)

Vijayababu Sriramulu

DB2 V7 Z/oS : Archive Log MODE(QUIESCE)
Dear List Users,

We have a daily job running in production, taking a system wide quiesce with the archive log command. We have a wait time of 300 (this is tuned based on the commit frequencies of our DB2 batch framework) with WAIT(YES) option. This job has been failing for the last 4/5 days since it is unable to drain the writers. This has been happening in the past and to fix this in the past, we had 4 steps in the job executing the same db2 command. If step1 fails, step2 tries it and it goes till step4 or till it succeeds. Each step has a wait time of 300 to drain all the writers.

Everytime when this job fails, there are UR's (count is not the same). Some times the UR's are 2, sometimes 3 and it varies between 2 and 6. When the archive log command goes to a wait mode, i did check for any hanging INDOUBT thread and there were NONE.

Here are my questions.

1. I cannot increase the wait time, since i am not sure if our db2 batch jobs have will get timed out for waiting for too long. The manual says that the drained threads are suspended until the quiesce comes out of wait. But is there any DB2 defined parameter which will knock out my db2 batch/distributed threads since this suspended time limit is too high. Like for example, if i increase the wait(300) to wait(500) tomorrow for example, those threads which are drained will be suspended for 500 seconds (at the worst case). Will there be any parameter which will knock out these drained threads since it waited for 500 seconds ?

2. Is there any way i can find out the UR's that are unable to drain the locks. The failure message from ARCHIVE LOG MODE(QUIESCE) says active UR's = 2. Is there anyway i can find out which are the active UR's at this point of time. If not immediatley, is it possible to get this information the next day from my SMF records. We have DB2 PM for extracting reports from SMF records. If yes, any specific trace has to be switched on in this case ?

3. Apart from the above 2 approaches, is there any other way i can make this job run successfully ?

Your suggestions/clues/hints/directions will be of great help to me.

Thanks in advance for your time.



Welcome to the IDUG DB2-L list. To unsubscribe, go to the archives and home page at http://www.idugdb2-l.org/archives/db2-l.html. From that page select "Join or Leave the list". The IDUG DB2-L FAQ is at http://www.idugdb2-l.org. The IDUG List Admins can be reached at [login to unmask email] Find out the latest on IDUG conferences at http://conferences.idug.org/index.cfm