| Author |
Message
|
| aditya.aggarwal |
 Posted: Tue Jan 13, 2009 9:31 pm Post subject: put disable enable on a cluster queue |
 |
|
Master
Joined: 13 Jan 2009
Posts: 252
|
We have four queue manager namely QM1, QM2 , QM3, QM4 in to a MQ Cluster. QM1, QM2 are Full repository queue manager.
QM3, QM4 are partial repository queue manager. Qm3, QM4 have same local queue with name QUEUE.USE , which is part of MQ cluster also.
This cluster setup was working fine initially and message were arriving in the QUEUE.USE at both queue manager QM3 and QM4.
Recently we have stopped all Cluster Channels and after that put disabled the QUEUE.USE at both Queue Manager QM3 and QM4 for 3 hours.
After sometime we have enabled the put call at QUEUE.USE only at Queue Manager QM3. Keeping Put disabled at QUEUE.USE at Queue Manager QM4.
After this We have restarted Cluster channels.
Now we are facing big difficulties at the QUEUE.USE. All message at the queue manager QM3 are going in to DEAD Letter queue, even we have put enabled at the queue QUEUE.USE at Queue Manager QM3.
When I checked the queue properties it is showing me unexpected results.
At QM3:
dis ql(QUEUE.USE)
1 : dis ql(QUEUE.USE)
AMQ8409: Display Queue details.
DESCR(Local Queue for feed from ) PROCESS( )
BOQNAME( ) INITQ( )
TRIGDATA( ) CLUSTER(QM_CLUSTER)
CLUSNL( ) QUEUE(QUEUE.USE)
CRDATE(2008-11-09) CRTIME(21.47.56)
ALTDATE(2009-01-13) ALTTIME(14.53.47)
GET(ENABLED) PUT(ENABLED)-----
DEFPRTY(0) DEFPSIST(NO)
MAXDEPTH(500000) MAXMSGL(4194304)
BOTHRESH(1) SHARE
DEFSOPT(SHARED) HARDENBO
MSGDLVSQ(FIFO) RETINTVL(999999999)
USAGE(NORMAL) NOTRIGGER
TRIGTYPE(FIRST) TRIGDPTH(1)
TRIGMPRI(0) QDEPTHHI(80)
QDEPTHLO(20) QDPMAXEV(DISABLED)
QDPHIEV(DISABLED) QDPLOEV(DISABLED)
QSVCINT(999999999) QSVCIEV(NONE)
DISTL(NO) NPMCLASS(NORMAL)
DEFTYPE(PREDEFINED) TYPE(QLOCAL)
SCOPE(QMGR) DEFBIND(NOTFIXED)
IPPROCS(1) OPPROCS(0)
CURDEPTH(0)
dis qc(QUEUE.USE) all
16 : dis qc(QUEUE.USE) all
AMQ8409: Display Queue details.
DESCR(Local Queue for feed from ) CLUSTER(QM_CLUSTER)
QUEUE(QUEUE.USE) CLUSQMGR(QM3)
QMID(QM3_2008-09-13_12.56.06) CLUSDATE(2008-11-09)
CLUSTIME(20.47.56) ALTDATE(2009-01-12)
ALTTIME(09.54.16) CLUSQT(QLOCAL)
TYPE(QCLUSTER) PUT(DISABLED)--
DEFPRTY(0) DEFPSIST(NO)
DEFBIND(NOTFIXED)
AMQ8409: Display Queue details.
DESCR(Local Queue for feed from ) CLUSTER(QM_CLUSTER)
QUEUE(QUEUE.USE) CLUSQMGR(QM4)
QMID(QM4_2008-09-15_14.00.46) CLUSDATE(2008-11-17)
CLUSTIME(10.57.27) ALTDATE(2008-12-05)
ALTTIME(14.21.13) CLUSQT(QLOCAL)
TYPE(QCLUSTER) PUT(DISABLED)
DEFPRTY(0) DEFPSIST(NO)
DEFBIND(NOTFIXED)
at QM QM4:
dis ql(QUEUE.USE)
6 : dis ql(QUEUE.USE)
AMQ8409: Display Queue details.
DESCR(Local Queue for feed from ) PROCESS( )
BOQNAME( ) INITQ( )
TRIGDATA( ) CLUSTER(QM_CLUSTER)
CLUSNL( ) QUEUE(QUEUE.USE)
CRDATE(2008-11-09) CRTIME(23.03.46)
ALTDATE(2009-01-13) ALTTIME(15.08.49)
GET(ENABLED) PUT(DISABLED)
DEFPRTY(0) DEFPSIST(NO)
MAXDEPTH(500000) MAXMSGL(4194304)
BOTHRESH(1) SHARE
DEFSOPT(SHARED) HARDENBO
MSGDLVSQ(FIFO) RETINTVL(999999999)
USAGE(NORMAL) NOTRIGGER
TRIGTYPE(FIRST) TRIGDPTH(1)
TRIGMPRI(0) QDEPTHHI(80)
QDEPTHLO(20) QDPMAXEV(DISABLED)
QDPHIEV(DISABLED) QDPLOEV(DISABLED)
QSVCINT(999999999) QSVCIEV(NONE)
DISTL(NO) NPMCLASS(NORMAL)
DEFTYPE(PREDEFINED) TYPE(QLOCAL)
SCOPE(QMGR) DEFBIND(NOTFIXED)
IPPROCS(1) OPPROCS(0)
CURDEPTH(0)
dis qc(QUEUE.USE) all
3 : dis qc(QUEUE.USE) all
AMQ8409: Display Queue details.
DESCR(Local Queue for feed from ) CLUSTER(QM_CLUSTER)
QUEUE(QUEUE.USE) CLUSQMGR(QM3)
QMID(QM3_2008-09-13_12.56.06) CLUSDATE(2008-11-17)
CLUSTIME(10.57.27) ALTDATE(2008-11-17)
ALTTIME(10.12.46) CLUSQT(QLOCAL)
TYPE(QCLUSTER) PUT(ENABLED)
DEFPRTY(0) DEFPSIST(NO)
DEFBIND(NOTFIXED)
AMQ8409: Display Queue details.
DESCR(Local Queue for feed from ) CLUSTER(QM_CLUSTER)
QUEUE(QUEUE.USE) CLUSQMGR(QM4)
QMID(QM4_2008-09-15_14.00.46) CLUSDATE(2008-11-09)
CLUSTIME(22.03.46) ALTDATE(2008-12-05)
ALTTIME(14.21.13) CLUSQT(QLOCAL)
TYPE(QCLUSTER) PUT(DISABLED)
DEFPRTY(0) DEFPSIST(NO)
DEFBIND(NOTFIXED)
I am not able to understand why the QUEUE.USE is still showing put disabled at queue manager QM3 even I have put enabled it at QM3. Apart from that it is showing as put enable for QUEUE.USE of QM3 at queue manager QM4.
I need to resolve this problem also as all the message are going in to DLQ with error MQRC 2268: MQRC_CLUSTER_PUT_INHIBITED
I doubt that Full repository queue manager is not updating the changes of partial repository queue manager QM3.
Any help or suggestion to resolve this is much awaited and appreciated.
QM1, QM2 : z/os MQV5.3 queue manager
QM3, QM4: Solaris MQ V5.3 queue Manager.
Cheers
AA |
|
| Back to top |
|
 |
| masteringmq |
| Posted: Tue Jan 13, 2009 10:35 pm Post subject: |
|
|
Master
Joined: 20 Oct 2008
Posts: 200
|
| Did you alter PUT(ENABLED) attribute for QM3 or QM4?. Also did you run a refresh cluster from a queue manager to discard all locally held info about a cluster?. |
|
| Back to top |
|
|
| Mr Butcher |
| Posted: Tue Jan 13, 2009 10:50 pm Post subject: |
|
|
Padawan
Joined: 23 May 2005
Posts: 1716
|
| Quote: |
| I doubt that Full repository queue manager is not updating the changes of partial repository queue manager QM3. |
if you stop the lcuster channels, and then change queues that are in the cluster on a PR, the change information can not flow from the PR to the FT queuemanagers. Cluster channels do not only transfer the messages, they also transfer cluster related infiormation. The information may have reached the FR after you start the channels, but you did several changes during that channel - stop period, that may be the possible reason for things getting ojut of sync (5.3 version is not "that new", it also depends which CSD you have implemented)
before doing cluster refrehs, i would just repeat the changes, but this time with running cluster channels and then check if the information of the cluster queues put attribute is fine again.
_________________
Regards, Butcher |
|
| Back to top |
|
|
| aditya.aggarwal |
| Posted: Tue Jan 13, 2009 11:17 pm Post subject: |
|
|
Master
Joined: 13 Jan 2009
Posts: 252
|
Many Thanks Butcher & masteringmq.
We have did the same changes and Refersh cluster with running Cluster channel yesterday, but it does not resolved the issue. We are using CSD11.
We have altered the put(ENABLED) only at QM3..
Need more advise on this issue. Is it a bug with MQ V5.3 CLuster?  |
|
| Back to top |
|
|
| Vitor |
| Posted: Wed Jan 14, 2009 1:00 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| aditya.aggarwal wrote: |
| We have did the same changes and Refersh cluster with running Cluster channel yesterday, but it does not resolved the issue. |
Which queue manager did you issue the refresh against? Also you need all the cluster channels running or the repositories will remain out of sync.
| aditya.aggarwal wrote: |
| Is it a bug with MQ V5.3 CLuster? |
Not one I've experienced. Which is not a reason not to upgrade to v6.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| Vitor |
| Posted: Wed Jan 14, 2009 1:00 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
Moved to Clustering section
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| aditya.aggarwal |
| Posted: Wed Jan 14, 2009 2:11 am Post subject: |
|
|
Master
Joined: 13 Jan 2009
Posts: 252
|
Refersh at the Full Repositry Queue Manager...
problem is still same... |
|
| Back to top |
|
|
| aditya.aggarwal |
| Posted: Wed Jan 14, 2009 2:20 am Post subject: |
|
|
Master
Joined: 13 Jan 2009
Posts: 252
|
my SYSTEM.CLUSTER.COMMAND.QUEUE is showing the 30 messages pending in the queue.
is there any problem in the MQ Cluster.. I think that Full repositry queue manger QM1 is not able to receive the update from the QM3..? |
|
| Back to top |
|
|
| Vitor |
| Posted: Wed Jan 14, 2009 2:20 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
And you're certain that all the cluster channels are running? Not just the ones between QM3 and it's FR?
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| Vitor |
| Posted: Wed Jan 14, 2009 2:22 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
Slightly off topic, but what prompted you to stop the cluster channels in the first place? I'm guessing (given the put disabling) you were trying to halt message flow?
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| Mr Butcher |
| Posted: Wed Jan 14, 2009 2:48 am Post subject: |
|
|
Padawan
Joined: 23 May 2005
Posts: 1716
|
| Quote: |
| my SYSTEM.CLUSTER.COMMAND.QUEUE is showing the 30 messages pending in the queue |
then something is wrong with your cluster repository manager. the commands you issue to the cluster, of the information that the cluster is sending internally is not being processed.
check the amq error logs, check for fdc files, check for the proper process iof it is running (don't have the name at hand, but you'll find it easily in the docs or here)
_________________
Regards, Butcher |
|
| Back to top |
|
|
| exerk |
| Posted: Wed Jan 14, 2009 2:57 am Post subject: |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
And refreshing a cluster should only be done under direction from IBM support according to the manual.
And I'm with Vitor on this - why hard-stop the channels?
_________________
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 |
|
|
| aditya.aggarwal |
| Posted: Wed Jan 14, 2009 4:07 am Post subject: |
|
|
Master
Joined: 13 Jan 2009
Posts: 252
|
We have stopped the channel as we are doing migration of application from one platform to other platform.
We are able to resolve the issue at QM4 after boucing it and Restarting Cluster Channels at QM4. 
Issue is still same on QM3.. |
|
| Back to top |
|
|
| Vitor |
| Posted: Wed Jan 14, 2009 4:15 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| aditya.aggarwal wrote: |
| We have stopped the channel as we are doing migration of application from one platform to other platform. |
Again I ask why? Put disable the queues yes, but what was stopping the channels supposed to do? It's not like they'd stop messages flowing.
| aditya.aggarwal wrote: |
We are able to resolve the issue at QM4 after boucing it and Restarting Cluster Channels at QM4.
Issue is still same on QM3.. |
And again I ask if all the channels are running. Also if you've fixed that problem on the FR so the updates are being processed.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| aditya.aggarwal |
| Posted: Wed Jan 14, 2009 4:28 am Post subject: |
|
|
Master
Joined: 13 Jan 2009
Posts: 252
|
Thanks to all.
If we have not stopped the channels then the message have been flown to SYSTEM.DEAD.LETTER.QUEUE. That's why we have stopped the channels.
Issue is little bit differnt at QM3 now..
now i am getting the strange o/p after restarting the queue manager QM3 and its cluster chanels.
dis clusqmgr(*)
1 : dis clusqmgr(*)
AMQ8441: Display Cluster Queue Manager details.
CLUSQMGR(QM3) CLUSTER(QM_CLUSTER)
CHANNEL(TO.QM3.QM_CLUSTER)
AMQ8441: Display Cluster Queue Manager details.
CLUSQMGR(SYSTEM.TEMPQMGR.BROKERDEMO1.NET(512000))
CLUSTER(QM_CLUSTER) CHANNEL(TO.QM1.QM_CLUSTER)
start chl(TO.QM2.QM_CLUSTER)
2 : start chl(TO.QM2.QM_CLUSTER)
AMQ8227: Channel TO.QM2.QM_CLUSTER not found.
dis chs(TO.QM1.QM_CLUSTER)
3 : dis chs(TO.QM1.QM_CLUSTER)
AMQ8417: Display Channel Status details.
CHANNEL(TO.QM1.QM_CLUSTER) XMITQ(SYSTEM.CLUSTER.TRANSMIT.QUEUE)
CONNAME(BROKERDEMO1.NET(512000))
CURRENT CHLTYPE(CLUSSDR)
STATUS(RUNNING) RQMNAME(QM1)
Any clue on this? |
|
| Back to top |
|
|
|
|
