| |
|
RSS Feed - WebSphere MQ Support
|
RSS Feed - Message Broker Support
|
  |
|
| shared queues? |
« View previous topic :: View next topic » |
| Author |
Message
|
| boonus |
 Posted: Wed Jul 25, 2007 1:26 am Post subject: shared queues? |
 |
|
Newbie
Joined: 27 Sep 2006
Posts: 7
|
Hi,
I have 2 gmgrs(MQ1 and MQ2) in a cluster, full repos, running on WebSphere MQ for Linux. In both qmgrs I have a local queues "MESSAGES"
dis qcluster(*) all
27 : dis qcluster(*) all
AMQ8409: Display Queue details.
QUEUE(MESSAGES) TYPE(QCLUSTER)
ALTDATE(2007-07-24) ALTTIME(16.46.57)
CLUSDATE(2007-07-24) CLUSTER(MYCLUSTER)
CLUSQMGR(MQ1) CLUSQT(QLOCAL)
CLUSTIME(11.28.35) CLWLPRTY(0)
CLWLRANK(0) DEFBIND(OPEN)
DEFPRTY(0) DEFPSIST(NO)
DESCR( ) PUT(ENABLED)
QMID(MQ1_2007-07-24_10.28.35)
AMQ8409: Display Queue details.
QUEUE(MESSAGES) TYPE(QCLUSTER)
ALTDATE(2007-07-24) ALTTIME(16.44.25)
CLUSDATE(2007-07-24) CLUSTER(MYCLUSTER)
CLUSQMGR(MQ2) CLUSQT(QLOCAL)
CLUSTIME(12.00.03) CLWLPRTY(0)
CLWLRANK(0) DEFBIND(OPEN)
DEFPRTY(0) DEFPSIST(NO)
DESCR( ) PUT(ENABLED)
QMID(MQ2_2007-07-24_10.30.47)
The cluster handles workload and easier administration. But what if one of the machines goes down? Is there any common solution for this? I've read the "WebSphere MQQueue Manager Clusters" redbook and found something about cluster sharing with z/os. Is there any method I can accomplish this(or an other solution?) on MQ/Linux?
Thanks |
|
| Back to top |
|
 |
| Vitor |
| Posted: Wed Jul 25, 2007 1:51 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
If you have a look through the forum you'll find a lot of discussion around using MQ as a high availability tool to deal with outages; bottom line is it can't (or perhaps more accurately shouldn't) be used like this.
If a queue manager (even a clustered one) goes down then messages delivered to that queue manager but unprocessed will be inaccessable until that queue manager (or an HA failover of it) comes back up. The accepted term for this is "stranded messages" and you'll find discussion of that topic in here as well.
The cluster will, however, detect the outage and stop sending more messages to the failed qmgr; I'm talking about the ones already sent but unprocessed here.
The exception to this is where a z/OS qmgr is part of a queue sharing group as well as or instead of a cluster; this is specific to that platform, there's nothing like that in the Linux version.
If you're worried about a queue manager going down, implement a Linux HA solution and protect it.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| 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
|
|
|
|
