| Author |
Message
|
| kats |
 Posted: Tue Dec 19, 2006 11:17 am Post subject: Channel toggling between initialising/binding. |
 |
|
Voyager
Joined: 20 Apr 2006
Posts: 78
|
Platform : Sun Solaris. MQ 5.1 
We have 4 qmgrs running on this TEST SERVER
There was a space issue with /var/mqm" server so folks moved some FDC in "/var/mqm/errors" to "/tmp/XXX" dir.(I know they should have read it, but now even they are deleted)
After this, testers complained that msg are not flowing.
When we tried checking relevant qmgr with cmd "runmqsc QmgrName", MQ was not detecting the qmgr itself though it was up and running.
Then we killed all MQ processes, cleared memory/semaphores and restarted.
Now the issue is that One of the sdr chl is continously toggling between INITIALIZING, BINDING status and in process filling up the error log in "/var/mqm/errors". Also lot of FDC files are getting created in "/var/mqm/errors".
The RCVR chl is running.
Checked with Network guy, no issues with network.
There's no error in log files in "/var/mqm/qmgrs/QmgrName/errors" area.
Next we recycled the Qmgr , no luck, then we reboot the Server. No luck. Then I used saveqmgr utility, deleted the qmgr and recreated, still at same spot.
Are PTF's available for MQ 5.1(though I searched...couldn't find)
| MQSeries First Failure Symptom Report |
| ===================================== |
| |
| Date/Time :- Tuesday December 19 10:22:48 EST 2006 |
| Host Name :- eqndmdev1 |
| PIDS :- 5765B75 |
| LVLS :- 510 |
| Product Long Name :- MQSeries for Sun Solaris (Sparc) |
| Vendor :- IBM |
| Probe Id :- XC130003 |
| Application Name :- MQM |
| Component :- xehExceptionHandler |
| Build Date :- May 12 1999 |
| UserID :- 00008213 (adpndm)  This should be user mqm and not adpndm....isn't it!!!! |
| Program Name :- runmqchl_nd |
| Process :- 00007492 |
| Thread :- 00000001 |
| QueueManager :- U06151100 |
| Major Errorcode :- STOP |
| Minor Errorcode :- OK |
| Probe Type :- HALT6109 |
| Probe Severity :- 1 |
| Probe Description :- AMQ6109: An internal MQSeries error has occurred. |
| Arith1 :- 11 b |
|
errors from /var/mqm/errors:
12/19/06 10:37:12
AMQ6184: An internal MQSeries error has occurred on queue manager U06151100.
EXPLANATION:
An error has been detected, and the MQSeries error recording routine has been
called. The failing process is process 10759.
ACTION:
Use the standard facilities supplied with your system to record the problem
identifier, and to save the generated output files. Contact your IBM support
center. Do not discard these files until the problem has been resolved.
-------------------------------------------------------------------------------
12/19/06 10:37:13
AMQ6109: An internal MQSeries error has occurred.
EXPLANATION:
An error has been detected, and the MQSeries error recording routine has been
called.
ACTION:
Use the standard facilities supplied with your system to record the problem
identifier, and to save the generated output files. Contact your IBM support
center. Do not discard these files until the problem has been resolved.
Error from va/mqm/qmgrs/QmgrName/errors:
12/19/06 10:42:33
AMQ9002: Channel program started.
EXPLANATION:
Channel program '06151100_DTCU0.C1' started.
ACTION:
None.
-------------------------------------------------------------------------------
12/19/06 10:42:34
AMQ9002: Channel program started.
EXPLANATION:
Channel program '06151100_DTCU0.C1' started.
ACTION:
None.
-------------------------------------------------------------------------------
12/19/06 10:42:34
AMQ9002: Channel program started.
EXPLANATION:
Channel program '06151100_DTCU0.C1' started.
ACTION:
None.
-------------------------------------------------------------------------------
12/19/06 10:42:35
AMQ9002: Channel program started.
EXPLANATION:
Channel program '06151100_DTCU0.C1' started.
ACTION: |
|
| Back to top |
|
 |
| kats |
| Posted: Tue Dec 19, 2006 11:35 am Post subject: |
|
|
Voyager
Joined: 20 Apr 2006
Posts: 78
|
| A clue that I came across while searching this forum is that my listener(I've two listener on 1419 and 1420) may be running in asynchronous mode. How to check that listener actually running asynchronously! |
|
| Back to top |
|
|
| wschutz |
| Posted: Tue Dec 19, 2006 12:21 pm Post subject: |
|
|
Jedi Knight
Joined: 02 Jun 2005
Posts: 3316
Location: IBM (retired)
|
v5.1 with no csd's huh? No comment there .....
How are you starting the channel... as you point out:
UserID :- 00008213 (adpndm) This should be user mqm and not adpndm....isn't it!!!! |
Is adpndm at least in the mqm group?
_________________
-wayne |
|
| Back to top |
|
|
| JosephGramig |
| Posted: Tue Dec 19, 2006 12:33 pm Post subject: |
|
|
Grand Master
Joined: 09 Feb 2006
Posts: 1244
Location: Gold Coast of Florida, USA
|
You should also make sure that your QMGR will come up without any channels starting.
Make sure you do not use the inetd.conf to start the channels.
Start them with runmqlstr (when you are sure the QMGR started without errors).
And make sure mqm is the ID that starts everything to do with the QMGR processes.
PS: You know 5.1 is not supported, don't you?
_________________
Joseph
Administrator - IBM WebSphere MQ (WMQ) V6.0, IBM WebSphere Message Broker (WMB) V6.1 & V6.0
Solution Designer - WMQ V6.0
Solution Developer - WMB V6.1 & V6.0, WMQ V5.3 |
|
| Back to top |
|
|
| kats |
| Posted: Tue Dec 19, 2006 12:35 pm Post subject: |
|
|
Voyager
Joined: 20 Apr 2006
Posts: 78
|
It's running now!
No, adpndm is NOT in mqm group.
A file 'scylog.log' from /var/mqm/log was owned by user ID adpndm.
Now I moved the file to a junk folder and gave mqsc command start channel(XXX.YYY) and channel is in runnnig state.
A new file scylog.log got created in /var/mqm/log folder under user ID mqm.
This channel uses security exit.
How we got into this problem: there are lot of people tinkering with same problem using two ID's. That's what my guess says.
But I don't know exactly:By starting which process under adpndm, this problem came into picture.!! |
|
| Back to top |
|
|
| wschutz |
| Posted: Tue Dec 19, 2006 12:47 pm Post subject: |
|
|
Jedi Knight
Joined: 02 Jun 2005
Posts: 3316
Location: IBM (retired)
|
| Quote: |
| By starting which process under adpndm, this problem came into picture.!! |
runmqtrm.....
Jose ... this is the sender end of the channel, I believe....
_________________
-wayne |
|
| Back to top |
|
|
| kats |
| Posted: Tue Dec 19, 2006 1:18 pm Post subject: |
|
|
Voyager
Joined: 20 Apr 2006
Posts: 78
|
Heck...I forgot to mention...Yes, this is SDR side of channel. I believe no body used runmqtrm... but we've triggering enabled on XMITQ.
Thanks Wschutz. Thanks JosephGramig. |
|
| Back to top |
|
|
| wschutz |
| Posted: Tue Dec 19, 2006 1:19 pm Post subject: |
|
|
Jedi Knight
Joined: 02 Jun 2005
Posts: 3316
Location: IBM (retired)
|
sorry... typo...I meant RUNMQCHI
_________________
-wayne |
|
| Back to top |
|
|
| JosephGramig |
| Posted: Tue Dec 19, 2006 1:21 pm Post subject: |
|
|
Grand Master
Joined: 09 Feb 2006
Posts: 1244
Location: Gold Coast of Florida, USA
|
 Yes, inetd.conf and runmqlstr have no effect on the sender channel.
But don't use inetd.conf to start channels anyway! 
_________________
Joseph
Administrator - IBM WebSphere MQ (WMQ) V6.0, IBM WebSphere Message Broker (WMB) V6.1 & V6.0
Solution Designer - WMQ V6.0
Solution Developer - WMB V6.1 & V6.0, WMQ V5.3 |
|
| Back to top |
|
|
| kats |
| Posted: Tue Dec 19, 2006 2:26 pm Post subject: |
|
|
Voyager
Joined: 20 Apr 2006
Posts: 78
|
Since the default channel initiator is started automatically when you start a queue manager. . So does that mean the problem is caused by :
Someone logged in as user ID 'adpndm' (or in other words the ID that doesn't exists in mqm group), Started the Qmgr. !!!!!
Not understanding.  |
|
| Back to top |
|
|
| wschutz |
| Posted: Tue Dec 19, 2006 3:11 pm Post subject: |
|
|
Jedi Knight
Joined: 02 Jun 2005
Posts: 3316
Location: IBM (retired)
|
| kats wrote: |
| Since the default channel initiator is started automatically when you start a queue manager. |
Is that true for V5.1? I honestly don't remember ...
_________________
-wayne |
|
| Back to top |
|
|
| kats |
| Posted: Tue Dec 19, 2006 3:52 pm Post subject: |
|
|
Voyager
Joined: 20 Apr 2006
Posts: 78
|
Yes.
I just went thru the IBM Intercommunication for MQ 5.1.
| Quote: |
" In V5.1 of MQSeries for AIX, Compaq Tru64 UNIX, HP-UX, OS/2 Warp, Sun
Solaris, and Windows NT, a channel initiator is started automatically and the
number of channel initiators that you can start is limited. The default limit is 3.
You can change this using MAXINITIATORS in the qm.ini file for AIX, HP-UX,
OS/2 Warp, and Sun Solaris, and in the registry for Windows NT.
The default channel initiator is started automatically when you start a queuemanager. Similarly, it is stopped automatically when a queue manager is stopped" |
|
|
| Back to top |
|
|
| Nigelg |
| Posted: Wed Dec 20, 2006 1:43 am Post subject: |
|
|
Grand Master
Joined: 02 Aug 2004
Posts: 1046
|
So, the problem was caused by your security exit, which you forgot to mention in the first post.
When posting an FFST, put in up to the MQM call stack, so that the failing function can be seen. In your case it would have been rriCALL_EXIT, i.e. the failure is in user exit code.
Also, the user name adnpdm is the user running the channel, not the user starting MQ.
_________________
MQSeries.net helps those who help themselves.. |
|
| Back to top |
|
|
|
|
