| Author |
Message
|
| jeevan |
 Posted: Fri Feb 15, 2008 11:08 am Post subject: Creation of cluster queue |
 |
|
Grand Master
Joined: 12 Nov 2005
Posts: 1432
|
I am migrating an pld systems on 5.3 win 2000 / solaris 9 to mq 6, win2003 and solaris 10.
It is a cluster. we are doing like this:
build PR
create mq object ( not channel)
build FR
create objects but not channel
My question is
the back end queue manager has cluster queue. while creating these before the actually the queue manager joins the cluster, is there any problem creaing these clustered queue?
it sounds a lot of work creating them as local queue first and change them to clsuter queue later on. I am just exporting the existing defintion and running it except the channels and repos.
any suggestions would be appreciate
thanks |
|
| Back to top |
|
 |
| fjb_saper |
| Posted: Fri Feb 15, 2008 11:14 am Post subject: Re: Creation of cluster queue |
|
|
Grand High Poobah
Joined: 18 Nov 2003
Posts: 20756
Location: LI,NY
|
Creating the queue as a cluster queue before the qmgr actually joins the cluster should not be a problem...
Enjoy 
_________________
MQ & Broker admin |
|
| Back to top |
|
|
| jefflowrey |
| Posted: Fri Feb 15, 2008 11:15 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
It is, I guess, more work to run DEF QL and then ALTER QL instead of just running DEF QL.
It doesn't matter whether you set CLUSTER() at DEF QL or not.
Clusters will be updated when the qmgr either joins the cluster or when new cluster definitions are made (either through DEF QL or through ALTER QL).
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
|
| jeevan |
| Posted: Fri Feb 15, 2008 11:27 am Post subject: |
|
|
Grand Master
Joined: 12 Nov 2005
Posts: 1432
|
| jefflowrey wrote: |
It is, I guess, more work to run DEF QL and then ALTER QL instead of just running DEF QL.
It doesn't matter whether you set CLUSTER() at DEF QL or not.
Clusters will be updated when the qmgr either joins the cluster or when new cluster definitions are made (either through DEF QL or through ALTER QL). |
thanks you very much jeff. This is that I though and did but my colleague thinks other way.
Thank you once again, |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Fri Feb 15, 2008 11:40 am Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7722
|
On a non clustered QM, go look at the SYSTEM.CLUSTER.REPOSITORY.QUEUE. See how it has 1 message already there, and the q is open by the CLuster Repository Manager? This QM is ready to join a cluster and thell the cluster about itself.
Now go to any local q on this non clustered QM and cluster the queueu. Then go look at SYSTEM.CLUSTER.REPOSITORY.QUEUE again. There will be a new message there that contains info about this new clustered queue. So when this QM finally joins the cluster it will tell the Full Repository about itself and the clustered queue.
Do NOT assume that there is a one to one relationship with the # of messages in SYSTEM.CLUSTER.REPOSITORY.QUEUE and the # of clustered objects on a QM.
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
| jeevan |
| Posted: Fri Feb 15, 2008 12:54 pm Post subject: |
|
|
Grand Master
Joined: 12 Nov 2005
Posts: 1432
|
| PeterPotkay wrote: |
On a non clustered QM, go look at the SYSTEM.CLUSTER.REPOSITORY.QUEUE. See how it has 1 message already there, and the q is open by the CLuster Repository Manager? This QM is ready to join a cluster and thell the cluster about itself.
Now go to any local q on this non clustered QM and cluster the queueu. Then go look at SYSTEM.CLUSTER.REPOSITORY.QUEUE again. There will be a new message there that contains info about this new clustered queue. So when this QM finally joins the cluster it will tell the Full Repository about itself and the clustered queue.
Do NOT assume that there is a one to one relationship with the # of messages in SYSTEM.CLUSTER.REPOSITORY.QUEUE and the # of clustered objects on a QM. |
Thank you very much Peter. I appreciate your detailed and clear explanation how it works. |
|
| Back to top |
|
|
|
|
