IBM Support

Ask the Experts Replay: MQ Clustering Gotchas! Avoiding Cluster Administration Pitfalls



IBM Ask the Experts discussion about MQ Clustering Gotchas! Avoiding Cluster Administration Pitfalls on 19 November 2013.
MQ Clustering was introduced in v5.1 and customers have adopted the clustering method of messaging for ease of admin. and workload balance. Certain rules apply for config and admin, which if not followed, cause adverse effects. This session discusses ways to avoid MQ Cluster administration mistakes.


Ask the Experts sessions are conducted in a question and answer format. The table below provides a time index (minutes:seconds) to the recording and describes the questions or topics discussed. You can fast forward to any question using the time index. A table containing the panel of experts is also included.

To play or download the audio of this Ask the Experts session, see the Audio Section of this document (audio not available until after the webcast).

See the Related Information Section of this document for a list of documents referenced during the presentation.

Ask the Experts session
19 November 2013 - 11:00 a.m. - 12:00 p.m. EST

Time Questions asked
00:00 General introduction
02:50 Technical introduction
04:00 Improperly removing a cluster queue manager from the cluster.
10:46 Improper use of Refresh Cluster and Reset Cluster administration commands.
18:04 Ignoring cluster related messages in the queue manager error logs.
21:50 Improperly using the MQ clustering feature for High Availability or Disaster Recovery purposes.
26:33 Improperly configuring MQ Cluster channels and repositories.
30:58 HIPER APAR IV25030
34:18 Open lines for live question and answer period
34:28 Have you had any reported issues with multi-instance queue managers? Should any special steps be taken when joining a multi-instance QM to a cluster?
37:24 Slide 17, Workload Balancing, how to troubleshoot imbalanced workload among the cluster?
39:26 Pertaining to HIPER APAR IV25030, would you only demote or refresh the full repositories if you're seeing this corruption?
43:27 Recently my servers moved data centers, when the queue managers and repositories came up in the new data center, their IP addresses were changed. What would be the best way to recover?
46:17 We run Queue Managers on a High Availability server, meaning that they may be started on different hosts. Can you explain why the Queue Managers shold not be shut down?
52:14 Clearing out the SYSTEM.CLUSTER.TRANSMIT.QUEUE, is there an updated Support Pack?
58:37 Imbalanced workload among Queue Managers in remote clusters, would destination sequence factors affect the load balancing?
01:03:41 We are running MQ on Solaris 10 and run into an issue that a new qmgr on another server can not join this cluster; the repository manager process keeps dying and messages in SYSTEM.CLUSTER.COMMAND.QUEUE are queing up. What could cause this issue?
01:06:43 Aside from the stranded messages, will other messages behind them also be hung pending delivery to other queue managers?
01:07:20 Is the administration error missing on slide 21?
01:07:43 Is there an updated support pack to remove stranded messages from SYSTEM.CLUSTER.TRANSMIT.QUEUE - the one referencd is for MQ V5.3?
01:08:48 what fixpack level would be EXPOSED to the problem in IV25030?
01:09:43 Slide 21: Create a new channel to a full repository queue manager. Is this done on the queue manager with the partial repository?
01:10:12 When adding Partial Repositories to a cluster, what are the considerations for which Full Repository we should aim the manual CLUSSNDR at? Is it fine to aim all Partial Repositories' manual CLUSSNDRs to one Full Repository? The partial will immediately create automatic cluster senders to both fulls anyway, so is it irrelevant which FR the partial aims its manual CLUSSNDR at?
01:11:29 I have two QM ids (with same QM) showing in full repos. Old QM id still exists even after the QM is taken out of cluster. All i need to is to reset the cluster to remove the QM id every time. How do i make the old QM ID (with the same QM) to go automatically?
01:12:35 Are there any best practices or considerations you can post with regards to DR setups? For instance, when bringing up Full Repo's in DR mode (for testing purposes). Should we rely solely on network seperation, etc.?
01:13:46 Closing remarks
01:14:51 End of Call

Panel of Experts:
Andy Emmett WebSphere MQ Distributed L3 Support
Yana Johnson WebSphere MQ Distributed L2 Support
Valerie Lampkin WebSphere MQ Distributed L2 Support
William (Bill) Moss Websphere MQ Distributed L2 Support
Miguel Rodriguez WebSphere MQ Distributed L2 Support



To replay the recording of this conference call, click on Download Audio (audio in MP3 format). Right-click and select Save As to store the file on your local computer for later playback. Remember that you can fast forward to any question using the time index.

Related information

WebSphere Support Technical Exchange

Cross reference information
Segment Product Component Platform Version Edition
Business Integration IBM MQ Clustering AIX, HP-UX, Linux, OpenVMS, Solaris, Windows 9.0, 8.0, 7.5, 7.1, 7.0

Document information

More support for: WebSphere MQ

Software version: 7.0, 7.1, 7.5, 8.0, 9.0

Operating system(s): AIX, HP-UX, Linux, OpenVMS, Solaris, Windows

Reference #: 7039756

Modified date: 20 November 2013

Translate this page: