| |
|
RSS Feed - WebSphere MQ Support
|
RSS Feed - Message Broker Support
|
| HELP: MQLQMAG problems on MQ 5.1, 4 CPU NSK Tandem S74000 |
« View previous topic :: View next topic » |
| Author |
Message
|
| LouML |
 Posted: Tue Nov 22, 2005 11:40 am Post subject: |
 |
|
Partisan
Joined: 10 Nov 2005
Posts: 305
Location: Jersey City, NJ / Bethpage, NY
|
| LuisFer wrote: |
0825 (Hex) = 2085 (Dec).
The permanent Queues are good, i always reference to Temp/Dynamic Queues.
The Local Queue MQ Message Agent (LQMQMAG) is called for a Local Application to perform the MQ Calls. A local aplication is (by example) the MQMC, PCFs.
I should open a PMR to IBM with this problem.
Sorry |
Thank you for your help, Luis.
I should say that this also happens with the MQMCACAL program as well.
We're still looking into this. For some reason, this only happens on our production system, not our development. Both systems are running the same MQ versions. |
|
| Back to top |
|
 |
| LouML |
| Posted: Tue Dec 20, 2005 11:34 am Post subject: |
|
|
Partisan
Joined: 10 Nov 2005
Posts: 305
Location: Jersey City, NJ / Bethpage, NY
|
Well, we finally figured it out. I should have paid closer attention to the zcpSystemMessage in the FFST. It turns out that the MQLQMAG process was receiving an unexpected system message.
We had some orphaned processes running for months. These processes had various randomly generated process names ($Xnnn, $Ynnn or $Znnn). The scenario was as follows.
EXAMPLE:
1 - Process $X123 is started and it has a hometerm of $Y456.
2 - For some reason, process $X123 is not stopped, and continues to run.
3 - The Hometerm TACL $Y456 is stopped at some later date and process name $Y456 is returned to the available process name pool.
4 - Some time later (days, weeks, months) process name $Y456 is selected as a random process name for the MQLQMAG program.
5 - For some reason $X123 decides to write a system message to $Y456 again.
6 - Process $Y456 (which is now the MQLQMAG program) does not understand what this message is and reports it to the error log.
We tracked this down with help from IBM. They helped us debug the MQLQMAG program when the error came up. We were able to get the processhandle and convert it to a process name. We noticed that each time it happened, the program that sent the message was the same type. Turns out there were over 100 instances of this orphaned program running. Since they've been stopped, we have yet to receive another error.
Now that we know what to look for, we are watching this particular program that is orphaned. When we see one, we do a PSTATE on it. In a few cases, we've seen the hometerm file open with a Last Error status of 201, whch proves our theory.
Thanks to all for you help! |
|
| Back to top |
|
|
| LuisFer |
| Posted: Tue Dec 20, 2005 10:01 pm Post subject: |
|
|
Partisan
Joined: 17 Aug 2002
Posts: 302
|
Congrats, it's a good thing to know it.
Regards |
|
| Back to top |
|
|
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
|
|
