|
RSS Feed - WebSphere MQ Support
|
RSS Feed - Message Broker Support
|
 |
|
WMQ V7 and Windows 2008 High Availablity MSCS Cluster issue |
« View previous topic :: View next topic » |
Author |
Message
|
JRome |
Posted: Thu Jul 28, 2011 2:57 am Post subject: WMQ V7 and Windows 2008 High Availablity MSCS Cluster issue |
|
|
Novice
Joined: 01 Jul 2011 Posts: 15
|
Hi I have an issue here can someone help me.
We have Two nodes having windows 2008 R2 as it's OS which are MS Clustered nodes.
There are 2 Queue Managers that are on each server nodes, but are currently writing to a local drive(1 drive on each) on each nodes.
Say Node N1 has M1 and M2 queue manager and is writing to D(local drive) (D:\EAI\IBM\WMQ\LOGS\M1\active... and D:\EAI\IBM\WMQ\LOGS\M2\active\..)
And D:\EAI\IBM\WMQ\M1\error ..ect, D:\EAI\IBM\WMQ\M2\error...ect.
The Same way Node N2 has to 2 Queue Managers say S1 and S2 writing it logs and data/files as same node N1 queue managers writing to A local disk ie: L:\EAI\IBM\WMQ\Logs\S1\active.. and L:\EAI\IBM\WMQ\logs\S2\active\..
and L:\EAI\IBM\WMQ\S1\error..ect, L:\EAI\IBM\WMQ\S2\error..ect.
The above is how it is currently and I need to make all 4 Queue Managers ie M1,M2, S1,S2 ACTIVE,ACTIVE High availability Queue managers (MSCS Clustered).
Say for instance firstly I want to make M1 and M2 Queue Managers HA (Highly available) and include in the MSCS Cluster.
Note That in a MSCS Clustered Environment M1 and M2 will have separate SAN Disks(E,F) to enable independent failovers i.e M1 can failover to Node N2 while M2 CONTINUES TO stays in Node N1.
Following is the methodology we follow which works for Win 2003 MSCS.
1) I set M1,M2 queue manager 'start up' property to manually start up and stop M1,M2 qmgrs ON NODE N1.
2) Stop the IBM MQ Series service (Under window services/Computer manager) on Node N1 where M1 and M2 are.
3) Run haregtyp /r
**** NOTES and Questions:
Let me know whether 1,2,3 order is correct
Step 3 will be run only once on Node N1.
Do I do Steps 1 on all 4 Queue Managers on Node-N1 and Node -N2 and step 2 per node, before I start doing Step 3 flowed by the rest of the steps ?.
******end Note and questions
4) RUN the following command on Node N1: hamvmqm /m M1 /dd E:\EAI\IBM\WMQ /ld E:\EAI\IBM\WMQ\LOGS\
5) RUN the following command on Node N1: hamvmqm /m M2 /dd F:\EAI\IBM\WMQ /ld F\:\EAI\IBM\WMQ\LOGS\
6) Create the MCSC instance on the MSCS Cluster for queue manager M1, M2, making E,F , NETWORK IP resource dependant on the qmgrs M1 and M2 respectively.
**** Pre requisite **************
Pre req 1) The SAN Disk is already in place (ie E,F in this case), and is part of the MSCS Cluster each having a dependency of it own network IP resource which is also included in the MSCS Cluster.
Pre req 2) The SAN disk and IP address resource failover and fail back are successful.
******PROBLEM Definition *************
We are having a huge problem here when it comes to WINDOWS 2008 R2 OS.
I am being told by the SAN Disk/Hardware Team that unless we have first
Created MSCS instance for the queue managers the SAN disks made available to the N1 node but not part of MSCS clustered, at this stage to do STEP-4 and STEP 5 after labeling the provisioned SAN (E, F).
Only when the above(last 3 lines on the above paragraph) is done, Windows 2008 R2 will allow the two SAN Disk to be included part of MSCS Cluster.
How will you implement MSDTC for this scenario, will you have in all for SANs on both the node or 1 per node ?.
******QUESTIONS*******
Is there a way on Windows 2008 R2 to follow the order step 1.. 6 as we did on Windows 2003 MSCS Clustering when we make queue managers Highly available ?.
Also Does Windows 2008 R2 allow dependencies (M1 QMGR DEPENDS ON E) to be put in place while administering MSCS CLUSTER.
or does it do this automatically ?.
******END Question******
7) Fail over fail back testing for M1 and M2 (From Node-N1 to N2 and Back.
******QUESTION*******
Do I start IBM WMQ series Windows Service on Node N1 before I do Step-7.
***End Question******
Please your answers are very valuable and appreciated, and also please point me to some use full documentation, especially the MS Clustering for WINDOW 2008 R2 how to implement active active QMGRs on each node for TWO queue manager on each node as given in the introduction at the top.
And how to include Provisioned SAN disks to be include in MSCS -HA on WINDOW 2008 R2. |
|
Back to top |
|
 |
exerk |
Posted: Thu Jul 28, 2011 3:21 am Post subject: |
|
|
 Jedi Council
Joined: 02 Nov 2006 Posts: 6339
|
Just follow the instructions within the Info Centre for creating/migrating queue managers to MSCS control, and as for your last question "...And how to include Provisioned SAN disks to be include in MSCS -HA on WINDOW 2008 R2..." your Windows Sys Admins will know how to set up the hardware.
Things to note:
1. Each queue manager should have its own disks;
2. Each queue manager and dependent resources, should run in an independent resource group;
3. Each queue manager should ideally have its own Virtual IP Address;
4. Consider settings affinities (preferred node) for each queue manager.[/list] _________________ 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 |
|
 |
|
|
 |
|
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
|
|
|
|