| |
|
RSS Feed - WebSphere MQ Support
|
RSS Feed - Message Broker Support
|
  |
|
| PR was down and made it up but it wont showup in FR |
« View previous topic :: View next topic » |
| Author |
Message
|
| chris boehnke |
 Posted: Thu Jun 26, 2008 4:07 pm Post subject: PR was down and made it up but it wont showup in FR |
 |
|
Partisan
Joined: 25 Jul 2006
Posts: 369
|
Hi Guys,
Here is my Cluster env setup with the 5 QMgrs in which 3 are FR and 2 PRs.
QM1, QM2 and QM3 are Full Repos
QM4 and QM5 are Partial Repos
QM1 is the primary Full Repo....QM2 and QM3 are assumed to be secondary Full Repos.
Here are definitions FOR FRs:
QM1:
CLUSRCVR: TO.QM1
CLUSSDR: TO.QM2
QM2:
CLUSRCVR: TO.QM2
CLUSSDR: TO.QM1
QM3:
CLUSRCVR: TO.QM3
CLUSSDR: TO.QM1
Here are the definitions for PRs:
QM4:
CLUSRCVR: TO.QM4
CLUSSDR: TO.QM1
QM5:
CLUSRCVR: TO.QM5
CLUSSDR: TO.QM1
All the Queues on QM4 are defined with priority(CLWLPRTY(5)) when we compare the queues on QM5 which are with priority zero(CLWLPRTY(0)).
That means when both the PR QMgrs are up and running, the msgs from FRs should sned msgs to QM4 queues and in case QM4 is down the msgs should be diverted to QM5.
Question:
case 1: When the QM4 goes down, the FRs are not updated immediately and they stull assume QM4 exists and the messages are not diverted to QM5.
I did a refresh cluster on all the FRs but no use. Then I restarted the FRs by suspending and resuming them. Then thy are knowing QM4 is down and QM5 is part of cluster.
case 2: When the QM4 comes up again, they are still sending the msgs to QM5.
Do I need to restart the FR QMgr by suspending and resumeing each time when the the PR goes down and again comes up?.
Why the FRs are not updating quickly when I issue SUSPEND QMGR CLUSTER(clusname)or RESUME QMGR CLUSTER(clusname)?.
Am I doing anything in the definitions or do I need to take any precautions while adding and removing aQMgr from a cluster?.
Thanks in advance. |
|
| Back to top |
|
 |
| sami.stormrage |
| Posted: Thu Jun 26, 2008 10:50 pm Post subject: answer |
|
|
Disciple
Joined: 25 Jun 2008
Posts: 186
Location: Bangalore/Singapore
|
It is a know problem with MQ. Try netprty instead of cluswlprty. Install HACMP if ur are finding it so troublesome.
Yes a cluster resfresh on the full repository takes some time for the information to get updated, since you do not have a direct communication b/w all the FR's and PR's. Irony is that, a cluster refresh does more harm than good, if u have accidently left some qmgr suspended or have large number of Q's
_________________
*forgetting everything  * |
|
| Back to top |
|
|
| sami.stormrage |
| Posted: Thu Jun 26, 2008 10:53 pm Post subject: |
|
|
Disciple
Joined: 25 Jun 2008
Posts: 186
Location: Bangalore/Singapore
|
Suspending and resuming ur PR is enough to sort out the thread connection problems..
_________________
*forgetting everything * |
|
| Back to top |
|
|
| chris boehnke |
| Posted: Fri Jun 27, 2008 9:57 am Post subject: Re: answer |
|
|
Partisan
Joined: 25 Jul 2006
Posts: 369
|
| sami.stormrage wrote: |
It is a know problem with MQ. Try netprty instead of cluswlprty. Install HACMP if ur are finding it so troublesome.
Yes a cluster resfresh on the full repository takes some time for the information to get updated, since you do not have a direct communication b/w all the FR's and PR's. Irony is that, a cluster refresh does more harm than good, if u have accidently left some qmgr suspended or have large number of Q's |
Where is the netprty property, whether its the Qmgr property or its the Queue property?. Could you let me know how HACMP can affect my cluster?. Thanks. |
|
| Back to top |
|
|
|
|
|
| |
|
Page 1 of 1 |
|
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
|
|
|
|
