MVSFORUMS.com

[linkingpark]

I know it occurs b'cuz of excessive Db2 requests. I want to more about it
What are all the possibilities that may lead to rldb?

[kolusu]

linkingpark,

You need to give us some more details. Is it something related to DB2 Locks?

Kolusu

[linkingpark]

let me tell you a situation when RLDB occured.

There is a program which is being triggered by a TD queue. The processing logic of the program is such that

when the first instance of the program is running and before it actually terminates, there are 4 records being put in the queue simultaneously. This results in 4 new instances of the program being created. But only the first instance will be effective in processing the records. Other instances will die off soon.

This is because of the program's processing logic.

As an initial step, the program checks for the status of the module.

-----> Only if it is in 'IDLE' state, the queue is read and processed (links to a DB2 intensive I/O program. This actually abended due to RLDB). After processing the fetched record, it checks to see if the queue has anymore records to be processed. If there is any, processes them . Otherwise, changes the status to 'IDLE' and terminates.

-----> If the program senses the module to be in ACTIVE state, it doesn't read the queue. It just terminates.

And most of all the syncpoint is set after each record is processed(when the control again returns to the linking pgm).
Does setting of syncpoint release the thread and re-initialize the DB2 count?

What could have been the possible reason for the abend?