| Author |
Message
|
| mq_population |
 Posted: Thu Jul 15, 2010 9:24 pm Post subject: AMQ6091 |
 |
|
Apprentice
Joined: 14 Jul 2009
Posts: 26
|
Hi All
We are facing an issue of which not very certain which should be best approach - and need some advice !
1. The error in MQ logs shows -
07/12/10 10:52:42
AMQ6091: An internal WebSphere MQ error has occurred.
EXPLANATION:
Private memory has detected an error, and is abending due to invalid head or
tail tag. The error data is 542640440.
2. Corresponding FDC shows -
WebSphere MQ First Failure Symptom Report |
| ========================================= | Product Long Name :- WebSphere MQ for AIX |
| Probe Id :- XC006001 |
| Application Name :- MQM |
| Component :- xcsFreeMem |
| CMVC level :- p530-13-L061205 |
| Build Type :- IKAP - (Production) |
| Major Errorcode :- xecS_I_PRIVATE_MEMORY_ERROR |
| Minor Errorcode :- OK |
| Probe Type :- MSGAMQ6091 |
| Probe Severity :- 4 |
| Probe Description :- AMQ6091: An internal WebSphere MQ error has occurred. |
| FDCSequenceNumber :- 0 |
| Comment1 :- invalid head or tail tag
We had opened a PMR with IBM - but It is MQ version 5.3 and CSD13 - and we dont have any Service Extension contract
Can anyone advice if they have faced such issues and what you did to resolve - or any other pointers from experts please?
I could not locate any relevant APAR.
Is an upgrade to V6 going to resolve
Thanks! |
|
| Back to top |
|
 |
| shashivarungupta |
| Posted: Thu Jul 15, 2010 10:05 pm Post subject: |
|
|
Grand Master
Joined: 24 Feb 2009
Posts: 1343
Location: Floating in space on a round rock.
|
|
| Back to top |
|
|
| shashivarungupta |
| Posted: Thu Jul 15, 2010 10:29 pm Post subject: |
|
|
Grand Master
Joined: 24 Feb 2009
Posts: 1343
Location: Floating in space on a round rock.
|
|
| Back to top |
|
|
| mq_population |
| Posted: Thu Jul 15, 2010 10:57 pm Post subject: |
|
|
Apprentice
Joined: 14 Jul 2009
Posts: 26
|
|
| Back to top |
|
|
| exerk |
| Posted: Thu Jul 15, 2010 11:14 pm Post subject: |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
| mq_population wrote: |
| ...This means if we upgrade to 6010 - it should be resolved correct ? |
Even if an upgrade doesn't solve the issue immediately, you will have moved to a supported version, in which case you can open another PMR and get IBM on the case 
_________________
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 |
|
|
| shashivarungupta |
| Posted: Thu Jul 15, 2010 11:30 pm Post subject: |
|
|
Grand Master
Joined: 24 Feb 2009
Posts: 1343
Location: Floating in space on a round rock.
|
| mq_population wrote: |
| This means if we upgrade to 6010 - it should be resolved correct ? |
Probe Id :- XC006001, when you search this string in the given ibm link you would find multiple fixes for that.. you gottu check error cause whether your mq has clustered setup or not, its because of exit or not , or because of ssl or something else. keep the kind of operating system in your mind as well while fixing it.
_________________
*Life will beat you down, you need to decide to fight back or leave it. |
|
| Back to top |
|
|
| sumit |
| Posted: Fri Jul 16, 2010 1:25 am Post subject: |
|
|
Partisan
Joined: 19 Jan 2006
Posts: 398
|
| mq_population wrote: |
| This means if we upgrade to 6010 - it should be resolved correct ? |
If upgrade/upversion is in your agenda, then go with WMQ7. Support for WMQ6 will end after an year.
With WMQ7, you can be on supported version for long.
_________________
Regards
Sumit |
|
| Back to top |
|
|
| Andyh |
| Posted: Thu Jul 29, 2010 4:52 am Post subject: AMQ6091 |
|
|
Master
Joined: 29 Jul 2010
Posts: 239
|
When MQ allocates private memory in a process it uses functions which wrapper malloc/free. The malloc wrapper (xcsGetMem) function adds a header and a trailer to the allocated storage, and the free wrapper (xcsFreeMem) checks the header and trailer are intact.
This FDC is raised when a header/trailer is found to be invalid, and typically signifies a storage overwrite of some kind.
Because this is simply private memory in the process address space then anything running in that process could be responsible for the
overwrite of the header/trailer.
A typical action upon receiving this FDC is to check what may have changed in the period leading to the failure. In the case of a failure in a cicsas process then candidates would include:
1. service applied to WMQ
2. service applied to CICS
3. any changes to CICS application code or exits
The two APARs mentioned previously in this post are specific to clustering and channel code and so are highly unlikely to effect a cicsas process. |
|
| Back to top |
|
|
|
|
