Peripheral rant:
And
when that happens, you have a stuck message (that requires manual intervention)
and a perpetual lag time (until the intervention rids the
message). To illustrate, suppose you have 1 message arriving every 5
minutes. The 1:00 trigger goes to the DLQ and the 1:00
application message doesn't get processed until the 1:05 trigger.
The 1:05 message doesn't get processed until the 1:10 trigger,
etc.
I
know, your thingy works 99.999% of the time. But by measure, 1
avoided trouble call in the lifetime of the app more than pays back
the cost of coding a loop.
But
even if you don't want to go to the trouble of doing that, you are still better
off with trigtype=first. Just make sure to check backout count to keep MQ
from looping on an abend.
regards,
Dennis
-----Original Message-----
From: Driscoll Tom - Princeton [mailto:[EMAIL PROTECTED]
Sent: Monday, November 29, 2004 5:40 AM
To: [EMAIL PROTECTED]
Subject: Trigger Messages to DLQ
From: Driscoll Tom - Princeton [mailto:[EMAIL PROTECTED]
Sent: Monday, November 29, 2004 5:40 AM
To: [EMAIL PROTECTED]
Subject: Trigger Messages to DLQ
This is a Microfocus COBOL program on HP-UX 11.11 MQSeries v5.3.
I have a triggered application with the triggertype set to EVERY. The triggered program is written to process one message and then end (it's a very low volume application). Everything runs fine, but periodically a trigger message gets written to the Dead Letter Queue with the associated initiation queue as the intended destination, and RUNMQTRM as the application. The reason code in the DLQ message header appears to be a hex 00000109 (??). I can't find a reason code that matches this.There are no associated error messages written to the AMQERR01.LOG or FDC entries. From a user perspective, everything is processing fine.
Any ideas ?
