| Author |
Message
|
| PeterPotkay |
 Posted: Thu Feb 14, 2013 12:17 pm Post subject: |
 |
|
Poobah
Joined: 15 May 2001
Posts: 7723
|
awe nuts, you're right. I was comparing the last line of the shown messages and they are different, but its because the copy and past shows a different amount of lines for each. I didn't catch that.
Sooo, this could be the same exact same message, with the same content, showing some unexplained Expiry value behaviour
BUT, its still possible the same message has been picked and put back to the queue, setting all the fields in the MQMD the same each time, including the Put Date and Put Time, which would reset the Expiry to 600 each time.
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
 |
| vishBroker |
| Posted: Thu Feb 14, 2013 1:00 pm Post subject: |
|
|
Centurion
Joined: 08 Dec 2010
Posts: 135
|
Hi Guys,
I won't be misleading you by fake problem. ( using two different messages).
I am here to get help not the test/tease you guys.
It is the same message. As pointout, just a different copy-paste.
By the way - I tried this on AIX and expiry worked properly.
So it is Windows thing.
Another point - I did check backout count. It was not increasing so possibility that the message was retried(re-put) in the queue is not there.
And, I was really glad to see all you experts chipping in for this. It helped me a lot to go through the comments.
Regards,
Vishnu. |
|
| Back to top |
|
|
| bruce2359 |
| Posted: Thu Feb 14, 2013 1:11 pm Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9475
Location: US: west coast, almost. Otherwise, enroute.
|
| vishBroker wrote: |
| By the way - I tried this on AIX and expiry worked properly. So it is Windows thing. |
So easy to blame Windoze. It could be that it's a WMQ version/release/modification-level thing. Or it's difference in hardware or workload thing.
_________________
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 |
|
|
| vishBroker |
| Posted: Thu Feb 14, 2013 1:35 pm Post subject: |
|
|
Centurion
Joined: 08 Dec 2010
Posts: 135
|
well, the setup was already there. ( I will blame it along with windows)
pun apart, I compared windows QM vs Unix QM. I do not see any difference.
Loadbalancig - These are standalone QMs. No cluster - no loadbalancing. ( lower environments)
Version/release are the same as well.
Regards,
Vishnu. |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Thu Feb 14, 2013 4:30 pm Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7723
|
| vishBroker wrote: |
Another point - I did check backout count. It was not increasing so possibility that the message was retried(re-put) in the queue is not there.
|
Re-putting the messages is not the same as backing out the message.
A backout will show evidence of itself because the MQMD's backout count will be incremented, which is not the case here according to your screen shots.
But getting a message, and then putting it back to the queue with another MQPUT, could be used to create the scenario, if it has the necessary authorizations to set the necessary options to set all the MQMD fields to make the 2 messages look the same.
"Its a Windows thing" is not a valid reason. Its fun to bash Windows, but lets get real. No one expects the Expiry counter to act crazy on Windows and can say that's acceptable.
vishBroker, have you opened a PMR to see what IBM says about this?
What specific version / service pack of Windows is this?
What specific version / fixpack of MQ?
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
| vishBroker |
| Posted: Tue Feb 19, 2013 9:47 am Post subject: |
|
|
Centurion
Joined: 08 Dec 2010
Posts: 135
|
I have opened sev-2 with IBM.
(Non-Prod Issue).
Haven't got any response from them yet. Will update once I get some.
Here are few details -
OS - details
+++++++++++++
OS - MS windows server 2003 R2
Enterprise Edition
Service pack 2
MQ version 7.0.1
-CMDLEVEL(701) |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Tue Feb 19, 2013 10:45 am Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7723
|
Issue the dspmqver command from the server to get the exact MQ version. MQ 7.0.1 is not specific enough.
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
| vishBroker |
| Posted: Tue Feb 19, 2013 11:07 am Post subject: |
|
|
Centurion
Joined: 08 Dec 2010
Posts: 135
|
Thanks.
Here is the o/p
+++
C:\ibmtemp>dspmqver
Name: WebSphere MQ
Version: 7.0.1.8
CMVC level: p701-108-120224
BuildType: IKAP - (Production)
C:\ibmtemp>
++++ |
|
| Back to top |
|
|
| vishBroker |
| Posted: Fri Mar 01, 2013 10:56 am Post subject: |
|
|
Centurion
Joined: 08 Dec 2010
Posts: 135
|
Well, just got this reply from IBM.
++++++++++++
This seems a bizarre problem where the timing mechanism being used by
MQ to keep time to a higher precision based on the clock speed of the
system is completely out of sync with the normal, system, time on the
system. This resulted in incorrect trace times in the traces and, more
importantly, incorrect calculation of message expiry times.
Similar symptoms have been previously reported - but only in virtual
machines - not indicated in this PMR.
.
the time errors indicate a
problem similar to APAR IC80980. The APAR is included in FP 7.0.1.9 so I
recommend a retest at that level.
Alternatively an interim fix at 7.0.1.8, fix:
7.0.1.8-WS-MQ-Windows-LAIC80980, can be found in prb #5250.
+++++++++++
Will apply the patch and test again. |
|
| Back to top |
|
|
| bruce2359 |
| Posted: Fri Mar 01, 2013 11:17 am Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9475
Location: US: west coast, almost. Otherwise, enroute.
|
Thanks for the update.
_________________
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 |
|
|
| vishBroker |
| Posted: Mon Mar 04, 2013 8:55 am Post subject: |
|
|
Centurion
Joined: 08 Dec 2010
Posts: 135
|
Well, it worked with new patch.
FYI - The patch overwrites amqxcs2.dll and amqxcs.pdb files.
This is the patch I applied and restarted the QMs
++
56162.L6Q.000.7.0.1.8-WS-MQ-Windows-LAIC80980
++ |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Mon Mar 04, 2013 11:30 am Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7723
|
Thank you for following thru with IBM and letting us know the results!
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
|
|
