| Author |
Message
|
| shashivarungupta |
 Posted: Fri May 08, 2009 1:53 am Post subject: Auto defined cluster sender channel and 2085 for a queue |
 |
|
Grand Master
Joined: 24 Feb 2009
Posts: 1343
Location: Floating in space on a round rock.
|
Hi,
In a situation, when application connect to the clustered queue manager and tried to open a queue to put the message, it gets the mqrc exception 2085. But it does not happen all the times. It happens intermittently.
In the cluster, there are 2 queue managers and between them there were auto defined cluster sender channels, now we have defined the cluster sender channels explicitly.
Now the issue is resolved that means applications do not get 2085 at all.
I have a doubt, why application was getting 2085 "sometimes" and why it was working fine with auto cluster sender channel.
It is confirmed that queue was shared in the cluster.
Thanks
_________________
*Life will beat you down, you need to decide to fight back or leave it. |
|
| Back to top |
|
 |
| Vitor |
| Posted: Fri May 08, 2009 1:59 am Post subject: Re: Auto defined cluster sender channel and 2085 for a queue |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| shashivarungupta wrote: |
| now we have defined the cluster sender channels explicitly. |
If you have a cluster sender channel pointing at a queue manager which is not defined as a full respository, you are flirting with danger.
If you've mistyped and you mean you've declared a point to point then fine, but it's a bit dubious.
| shashivarungupta wrote: |
| I have a doubt, why application was getting 2085 "sometimes" and why it was working fine with auto cluster sender channel. |
Because your cluster has a problem; possibly because you're defining channels without thought to the consequences and/or issuing reset cluster commands for no good reason (i.e. not because IBM told you to).
If you do have a cluster sender channel pointing at a partial repository, remove it. Then diagnose why the cluster isn't updating definitions properly and causing 2085 errors.
Details of cluster configuration, topology and versions of WMQ in use will help us offer advice.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| queuemanager |
| Posted: Fri May 08, 2009 4:28 am Post subject: |
|
|
Apprentice
Joined: 28 Nov 2006
Posts: 43
Location: Bangalore
|
| Code: |
| In the cluster, there are 2 queue managers and between them there were auto defined cluster sender channels |
If there are 2 queuemanagers in the cluster You need to define to define Cluster senders explicitly only. |
|
| Back to top |
|
|
| Vitor |
| Posted: Fri May 08, 2009 4:32 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| queuemanager wrote: |
| Code: |
| In the cluster, there are 2 queue managers and between them there were auto defined cluster sender channels |
If there are 2 queuemanagers in the cluster You need to define to define Cluster senders explicitly only. |
Well spotted that poster! 
I missed that in the original post. If it's a cluster of 2 queue managers only (and how pointless is that?) then both must be by definition full repositories, defined as full repositories and have manual cluster sender channels pointing to them.
If you didn't do that, your cluster was malformed and hence you were getting intermittant 2085 errors.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Fri May 08, 2009 4:44 am Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7722
|
If there are only 2 QMs in the cluster and you didn't specify the clussndr channels between them since both had to be FRs, you didn't have an MQ cluster to begin with.
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
| exerk |
| Posted: Fri May 08, 2009 4:49 am Post subject: |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
Differences in interpretation perhaps? I read this:
| Quote: |
| ...In the cluster, there are 2 queue managers and between them there were auto defined cluster sender channels... |
To mean that within the cluster, there are two queue managers that have auto-defined CLUSSDR's to each other, not that there are only two queue managers in the cluster.
Hopefully shashivarungupta will be along in a minute to clarify...
_________________
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 |
|
|
| WMBDEV1 |
| Posted: Fri May 08, 2009 4:54 am Post subject: |
|
|
Sentinel
Joined: 05 Mar 2009
Posts: 888
Location: UK
|
| exerk wrote: |
Differences in interpretation perhaps? I read this:
| Quote: |
| ...In the cluster, there are 2 queue managers and between them there were auto defined cluster sender channels... |
To mean that within the cluster, there are two queue managers that have auto-defined CLUSSDR's to each other, not that there are only two queue managers in the cluster.
|
And if this is the case, Vitors point:
| Vitor wrote: |
If you have a cluster sender channel pointing at a queue manager which is not defined as a full respository, you are flirting with danger.
|
becomes rather poignant  |
|
| Back to top |
|
|
| Vitor |
| Posted: Fri May 08, 2009 5:14 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
When I wrote that, I was somewhat stupidly assuming we were talking about a cluster. Rather than a group of queue managers broadly lumped together and beaten until they worked.
Or 2 queue managers just sitting next to each other like teenagers on a blind date.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| exerk |
| Posted: Fri May 08, 2009 5:20 am Post subject: |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
| Vitor wrote: |
| ...Or 2 queue managers just sitting next to each other like teenagers on a blind date... |
When I was a teenager they had to be blind for me to be in with a chance... 
_________________
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 |
|
|
| shashivarungupta |
| Posted: Mon May 11, 2009 1:23 am Post subject: |
|
|
Grand Master
Joined: 24 Feb 2009
Posts: 1343
Location: Floating in space on a round rock.
|
| exerk wrote: |
Differences in interpretation perhaps? I read this:
| Quote: |
| ...In the cluster, there are 2 queue managers and between them there were auto defined cluster sender channels... |
To mean that within the cluster, there are two queue managers that have auto-defined CLUSSDR's to each other, not that there are only two queue managers in the cluster.
Hopefully shashivarungupta will be along in a minute to clarify... |
There are not just the two queue managers in the cluster.
And the two queue managers about which I said, are the full repositories.
_________________
*Life will beat you down, you need to decide to fight back or leave it. |
|
| Back to top |
|
|
| exerk |
| Posted: Mon May 11, 2009 1:32 am Post subject: |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
| shashivarungupta wrote: |
| ...And the two queue managers about which I said, are the full repositories. |
Then your problems were self-inflicted, and were caused by not following the correct procedure, as laid down in the manual; It is a requirement to define explicit CLUSSDR's between FR's.
_________________
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 |
|
|
| shashivarungupta |
| Posted: Mon May 11, 2009 2:29 am Post subject: |
|
|
Grand Master
Joined: 24 Feb 2009
Posts: 1343
Location: Floating in space on a round rock.
|
Got it that point !!
In the cluster manual ..its mentioned as "When a queue manager has joined a cluster by making its initial CLUSSDR and CLUSRCVR definitions, it does not need to make any other definitions for channels to other queue managers in the cluster. MQSeries automatically makes cluster-sender channel definitions when they are needed. When both the cluster-sender end and the cluster-receiver end of a channel are defined, the channel is started. An auto-defined channel remains active until it is no longer
needed and is shut down using the normal disconnect-interval rules.
Auto-defined cluster-sender channels take their attributes from those specified in the corresponding cluster-receiver channel definition on the receiving queue manager. Even if there is a manually-defined cluster-sender channel, its attributes are automatically modified to ensure that they match those on the corresponding cluster-receiver definition. Beware, therefore, that you could, for example, define a CLUSRCVR without specifying a port number in the CONNAME parameter, whilst manually defining a CLUSSDR that does specify a port number. When the
auto-defined CLUSSDR replaces the manually defined one, the port number (taken from the CLUSRCVR) becomes blank. The default port number would be used and the channel would fail."
These were the two paragraphs.
Could you please tell me where is the description of auto cluster sender channel with respect to FR and PR.
_________________
*Life will beat you down, you need to decide to fight back or leave it. |
|
| Back to top |
|
|
| Vitor |
| Posted: Mon May 11, 2009 2:32 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| shashivarungupta wrote: |
| Could you please tell me where is the description of auto cluster sender channel with respect to FR and PR. |
That is the definition. The role of a queue manager in a cluster does not affect the definition of auto channels. It only affects the requirements as source or target of manually defined channels.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| exerk |
| Posted: Mon May 11, 2009 2:37 am Post subject: |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
The paragraph is poorly worded, but:
| Quote: |
| ...When a queue manager has joined a cluster by making its initial CLUSSDR and CLUSRCVR definitions... |
Well, they are not made automatically, and therefore have to be made explicitly.
However, it's pretty explicit in the Clustering Manual!
_________________
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 |
|
|
| akm.mohan |
| Posted: Mon Dec 28, 2009 1:37 pm Post subject: Auto-defined clus-sdr channel is not coming up |
|
|
Apprentice
Joined: 07 Oct 2008
Posts: 41
|
Hi all,
I have gone thru your replies to shashivarungupta's question. I understood how the cluster works. Thanks for that. Here I have a question in auto-definition of channels.
My Example:
CLUSTER: CLUSTER
QMGRS: QMGR1, QMT1 and QMT2
FullRepositories: QMT1 and QMT2
PR: QMGR1
Whenever I put the messages by issuing the amqsreq script from QMGR1, I can see the response time from only QMT1 not from QMT2.
I investigated and found that the cluster-sender chanenl is missing in QMGR1 called TO.QMT2.
As per the IBM manuals and your posts, if these 3 qmgrs are in one cluster, cluster sender channel(TO.QMT2) will be automatically created by the auto-definition and it takes their attributes from those specified in the corresponding cluster-rcvr channel definition on the receiving qmgr. But here I don't see the channel called TO.QMT2, even by issuing the dis chs(*) or (TO.QMT2).
If I define manually, it works fine...
Please help me to find out why its not coming up automatically or in other way..why auto definition is not working in this senario.
(clusterqueue defbind is set to notfixed)
Thanks in advance. |
|
| Back to top |
|
|
|
|
