| Author |
Message
|
| lenburt |
 Posted: Fri Sep 24, 2004 12:27 pm Post subject: Linear Logging Problems |
 |
|
Novice
Joined: 24 Sep 2004
Posts: 15
|
I'm support a large volume application using WMQ 5.3 CSD07 on Solaris. We are currently configured with CIRCULAR logging in produsction but would like to be able to recover damaged queues so we are testing out LINEAR. We seem to have a problem bringing our queue managers down (endmqm -p). The endmqm seems to hang waiting on the completion of some logging related processes. Is there something special that needs to be done on shutdown when using linear logging?
ps -ef | grep orbit_orb1_ce
mqm 16646 1 0 07:52:44 ? 0:00 endmqm -p orbit_orb1_ce
mqm 16629 1 0 07:51:54 ? 0:03 amqzxma0 -m orbit_orb1_ce
mqm 16632 16629 0 07:51:55 ? 0:00 amqhasmx orbit_orb1_ce /var/mqm
mqm 16631 16629 0 07:51:54 ? 0:00 amqharmx orbit_orb1_ce /var/mqm
orbit1 26764 26712 0 12:53:05 pts/10 0:00 grep orbit_orb1_ce |
|
| Back to top |
|
 |
| bower5932 |
| Posted: Fri Sep 24, 2004 12:29 pm Post subject: |
|
|
Jedi Knight
Joined: 27 Aug 2001
Posts: 3023
Location: Dallas, TX, USA
|
| You shouldn't have to do anything special. Are you getting any FDCs written? Do you have any WMQ applications running that didn't specify FAIL_IF_QUIESCING? |
|
| Back to top |
|
|
| lenburt |
| Posted: Fri Sep 24, 2004 12:35 pm Post subject: |
|
|
Novice
Joined: 24 Sep 2004
Posts: 15
|
I am fairly sure that there were no application processes running when i attempted the endmqm. Here is the header from an FDC....
WebSphere MQ First Failure Symptom Report |
| ========================================= |
| |
| Date/Time :- Friday September 24 07:56:58 PDT 2004 |
| Host Name :- orppap01 (SunOS 5. |
| PIDS :- 5724B4103 |
| LVLS :- 530.7 CSD07 |
| Product Long Name :- WebSphere MQ for Sun Solaris |
| Vendor :- IBM |
| Probe Id :- XC338001 |
| Application Name :- MQM |
| Component :- xehAsySignalHandler |
| Build Date :- May 27 2004 |
| CMVC level :- p530-07-L040527 |
| Build Type :- IKAP - (Production) |
| UserID :- 00001043 (orbit1) |
| Program Name :- endmqm |
| Process :- 00016646 |
| Thread :- 00000002 |
| Major Errorcode :- xecE_W_UNEXPECTED_ASYNC_SIGNAL |
| Minor Errorcode :- OK |
| Probe Type :- MSGAMQ6209 |
| Probe Severity :- 3 |
| Probe Description :- AMQ6109: An internal WebSphere MQ error has occurred. |
| FDCSequenceNumber :- 0 |
| Arith1 :- 2 2 |
| Arith2 :- 16646 4106 |
| Comment1 :- SIGINT |
| |
|
| Back to top |
|
|
| siliconfish |
| Posted: Fri Sep 24, 2004 12:55 pm Post subject: |
|
|
Master
Joined: 12 Aug 2002
Posts: 203
Location: USA
|
| why are you "-e" preemptive shutdown? have u tried controlled "-c" and then "-i" before using -e option? |
|
| Back to top |
|
|
| lenburt |
| Posted: Fri Sep 24, 2004 1:09 pm Post subject: |
|
|
Novice
Joined: 24 Sep 2004
Posts: 15
|
we first tried endmqm -c, then endmqm -i, and finally endmqm -p
I may need to kill -9 the logging processes which seem to be hanging. There seems to be plenty of free space on the log filesystem so i'm not sure what the problem is. |
|
| Back to top |
|
|
| siliconfish |
| Posted: Fri Sep 24, 2004 1:26 pm Post subject: |
|
|
Master
Joined: 12 Aug 2002
Posts: 203
Location: USA
|
| This probe id XC338001, doesn't indicate a problem with MQ processes, as bower5932 had mentioned, there must be some application holding connections to the queue manager. |
|
| Back to top |
|
|
| fjb_saper |
| Posted: Fri Sep 24, 2004 2:04 pm Post subject: |
|
|
Grand High Poobah
Joined: 18 Nov 2003
Posts: 20756
Location: LI,NY
|
If the restart of the qmgr fails check the offending pids being returned.
(Unix)
This should tell you the owner of the offending process and what application he/she is running.
Enjoy  |
|
| Back to top |
|
|
|
|
