AFAIK there is no built-in mechanism to do what you want.  We have a similar requirement and I had to build it in to the custom code that sits between MQ and CICS user transactions.  Specifically, our CICS transactions would occassionally encountered IDMS problems (e.g. locked records) and the transaction would abend, causing the MQ record to roll back.  Our service layer would trap the abend, search a table by program name and see if there's a retry entry for the given abend code.  If found, the entry would contain the number of retries to attempt, along with a value that indicates a delay time.  The service layer would issue a CICS DELAY before continuing with the abend flow.  The service layer (transaction) would then be reinvoked, getting the rolled-back message and reattempting the failed transaction.

Emile Kearns <emilek-WybPWJutzi2iaJFv3xFhBQ@public.gmane.org> Sent by: MQSeries List <MQSERIES-0lvw86wZMd9k/bWDasg6f+2wyY2g16FtwPuJ0ROkVbw@public.gmane.org> 04/02/2007 08:14 AM Please respond to MQSeries List <MQSERIES-0lvw86wZMd9k/bWDasg6f+2wyY2g16FtwPuJ0ROkVbw@public.gmane.org>

To MQSERIES-0lvw86wZMd9k/bWDasg6f+2wyY2g16FtwPuJ0ROkVbw@public.gmane.org cc Subject Backout retry time

Hi all,

One of my developers wants to know if it is possible to have a delay when reading a backed out message from a queue.

In other words, for some reason the message is backed onto the queue and the same process that tried taking it off the first time should have a delay and retry getting it off the queue again after a time period has expired.

List Archive - Manage Your List Settings - Unsubscribe

Instructions for managing your mailing list subscription are provided in the Listserv General Users Guide available at http://www.lsoft.com