| Author |
Message
|
| bkiran2020 |
 Posted: Wed Aug 28, 2013 6:29 pm Post subject: Dead queue |
 |
|
Master
Joined: 20 Jan 2011
Posts: 243
Location: US
|
The dead queue is stalled with messages due to MQRC unknown object 2085
But I can see the destination queue is present in the QMGR.
How can the dead queue is stalled with 2085 code.
 |
|
| Back to top |
|
 |
| fjb_saper |
| Posted: Wed Aug 28, 2013 8:43 pm Post subject: |
|
|
Grand High Poobah
Joined: 18 Nov 2003
Posts: 20756
Location: LI,NY
|
Have you tried running a DLQ handler with retry? What happens?
_________________
MQ & Broker admin |
|
| Back to top |
|
|
| Tibor |
| Posted: Wed Aug 28, 2013 10:04 pm Post subject: |
|
|
Grand Master
Joined: 20 May 2001
Posts: 1033
Location: Hungary
|
| You should check the creation time of that queue comparing the arriving time of the messages in the DLQ. I can assume queue was created later... |
|
| Back to top |
|
|
| bruce2359 |
| Posted: Thu Aug 29, 2013 4:15 am Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9469
Location: US: west coast, almost. Otherwise, enroute.
|
| Tibor wrote: |
| You should check the creation time of that queue comparing the arriving time of the messages in the DLQ. I can assume queue was created later... |
For what purpose?
_________________
I like deadlines. I like to wave as they pass by.
ב''ה
Lex Orandi, Lex Credendi, Lex Vivendi. As we Worship, So we Believe, So we Live. |
|
| Back to top |
|
|
| Tibor |
| Posted: Thu Aug 29, 2013 4:21 am Post subject: |
|
|
Grand Master
Joined: 20 May 2001
Posts: 1033
Location: Hungary
|
| bruce2359 wrote: |
| Tibor wrote: |
| You should check the creation time of that queue comparing the arriving time of the messages in the DLQ. I can assume queue was created later... |
For what purpose? |
Because the original post contains a question: How can the dead queue is stalled with 2085 code. |
|
| Back to top |
|
|
| bruce2359 |
| Posted: Thu Aug 29, 2013 5:16 am Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9469
Location: US: west coast, almost. Otherwise, enroute.
|
| Tibor wrote: |
| bruce2359 wrote: |
| Tibor wrote: |
| You should check the creation time of that queue comparing the arriving time of the messages in the DLQ. I can assume queue was created later... |
For what purpose? |
Because the original post contains a question: How can the dead queue is stalled with 2085 code. |
How does queue creation-time relate to 2085? How does queue creation-time help in diagnosing the underlying problem?
_________________
I like deadlines. I like to wave as they pass by.
ב''ה
Lex Orandi, Lex Credendi, Lex Vivendi. As we Worship, So we Believe, So we Live. |
|
| Back to top |
|
|
| Tibor |
| Posted: Thu Aug 29, 2013 6:03 am Post subject: |
|
|
Grand Master
Joined: 20 May 2001
Posts: 1033
Location: Hungary
|
OK, let's see an example:
- messages in the DLQ (filtering for destination and put date/time):
| Code: |
$ q -m RAPTOR -i DLQ -df3 | grep -iE "dest. queue|put date|put time"
MQSeries Q Program by Paul Clarke [ V4.5 Build:Sep 15 2006 ]
Connecting ...connected to 'RAPTOR'.
Dest. Queue :'IK_BATCH_IN.....................................'
Put Date :'20130711'
Put Time :'11232728'
Dest. Queue :'IK_BATCH_IN.....................................'
Put Date :'20130711'
Put Time :'11234957'
...
|
- info about destination queue (creation date/time):
| Code: |
$ runmqsc RAPTOR
5724-H72 (C) Copyright IBM Corp. 1994, 2011. ALL RIGHTS RESERVED.
Starting MQSC for queue manager RAPTOR.
dis qlocal(IK_BATCH_IN) crdate crtime
2 : dis qlocal(IK_BATCH_IN) crdate crtime
AMQ8409: Display Queue details.
QUEUE(IK_BATCH_IN) TYPE(QLOCAL)
CRDATE(2003-09-30) CRTIME(09.01.25) |
If the reason is 2085, and creation date/time is later than messages' put date/time, it offers a reasonable explanation for an MQRC_UNKNOWN_OBJECT_NAME error. (In my example, the Reason code was different, of course.) |
|
| Back to top |
|
|
| bruce2359 |
| Posted: Thu Aug 29, 2013 7:35 am Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9469
Location: US: west coast, almost. Otherwise, enroute.
|
I would think it more likely that the objectname specified in the MQOD was misspelled, or the resolved objectname really did not exist, or the app MQCONNected to the wrong qmgr.
An admin deleting a queue, then redefining it, seems an unlikely cause - but possible.
_________________
I like deadlines. I like to wave as they pass by.
ב''ה
Lex Orandi, Lex Credendi, Lex Vivendi. As we Worship, So we Believe, So we Live. |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Thu Aug 29, 2013 10:33 am Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7722
|
| bruce2359 wrote: |
| I would think it more likely that the objectname specified in the MQOD was misspelled |
The OP said the name is correct.
| bruce2359 wrote: |
| or the resolved objectname really did not exist |
The OP said the queue does exist.
| bruce2359 wrote: |
or the app MQCONNected to the wrong qmgr.
|
I don't think that applies...the scenario already involves the app connecting to another QM. If they were connecting to the same QM that hosted the queue, the app would get a failed MQPUT and nothing would go to the DLQ.
| bruce2359 wrote: |
| An admin deleting a queue, then redefining it, seems an unlikely cause - but possible. |
I think Tibor is suggesting the more likely scenario of the queue not existing, never have existed, but the app attempting to use it before the MQ Admin defined it for the first time based on their request to build it.
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
| bruce2359 |
| Posted: Thu Aug 29, 2013 12:36 pm Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9469
Location: US: west coast, almost. Otherwise, enroute.
|
| PeterPotkay wrote: |
| I think Tibor is suggesting the more likely scenario of the queue not existing, never have existed, but the app attempting to use it before the MQ Admin defined it for the first time based on their request to build it. |
A local app attempting to use a non-existent queue would result in a 2085, but not a msg in the DLQ.
An upstream app MQPUTting putting a down the network to a qmgr that failed to resolve the qname in the XQH would result in a msg to the DLQ with 2085.
_________________
I like deadlines. I like to wave as they pass by.
ב''ה
Lex Orandi, Lex Credendi, Lex Vivendi. As we Worship, So we Believe, So we Live. |
|
| Back to top |
|
|
| exerk |
| Posted: Thu Aug 29, 2013 1:10 pm Post subject: |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
| bruce2359 wrote: |
| PeterPotkay wrote: |
| I think Tibor is suggesting the more likely scenario of the queue not existing, never have existed, but the app attempting to use it before the MQ Admin defined it for the first time based on their request to build it. |
A local app attempting to use a non-existent queue would result in a 2085, but not a msg in the DLQ.
An upstream app MQPUTting putting a down the network to a qmgr that failed to resolve the qname in the XQH would result in a msg to the DLQ with 2085. |
Although the OP has not definitively stated the message originated from another queue manager it's probably a safe assumption and the scenario as stated by PeterPotkay is still then the most likely explanation.
_________________
It's puzzling, I don't think I've ever seen anything quite like this before...and it's hard to soar like an eagle when you're surrounded by turkeys. |
|
| Back to top |
|
|
| wmbwmq |
| Posted: Thu Aug 29, 2013 1:26 pm Post subject: |
|
|
Acolyte
Joined: 18 Jul 2011
Posts: 66
|
damn to the depths of OP coming up with a question and not intervening to resolve an ongoing all-the-way-to-death fight.... 
PS : I am that guy who is paranoid and following this thread.. |
|
| Back to top |
|
|
| bkiran2020 |
| Posted: Fri Aug 30, 2013 5:53 pm Post subject: |
|
|
Master
Joined: 20 Jan 2011
Posts: 243
Location: US
|
| bruce2359 wrote: |
I would think it more likely that the objectname specified in the MQOD was misspelled, or the resolved objectname really did not exist, or the app MQCONNected to the wrong qmgr.
An admin deleting a queue, then redefining it, seems an unlikely cause - but possible. |
I have checked the above mention thing but this is not the case
|
|
| Back to top |
|
|
| bruce2359 |
| Posted: Fri Aug 30, 2013 7:33 pm Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9469
Location: US: west coast, almost. Otherwise, enroute.
|
Then look at the MQMD of the dead message to see what application produced the message, and find out where it ran.
_________________
I like deadlines. I like to wave as they pass by.
ב''ה
Lex Orandi, Lex Credendi, Lex Vivendi. As we Worship, So we Believe, So we Live. |
|
| Back to top |
|
|
| Tibor |
| Posted: Sun Sep 01, 2013 11:31 pm Post subject: |
|
|
Grand Master
Joined: 20 May 2001
Posts: 1033
Location: Hungary
|
| Please post the MQDLH (dead-letter-header) content to the forum with an appropriate tool (mqmon, q, etc.) |
|
| Back to top |
|
|
|
|
