| |
|
RSS Feed - WebSphere MQ Support
|
RSS Feed - Message Broker Support
|
  |
|
| Migrating from HACMP to Multi Instance |
« View previous topic :: View next topic » |
| Author |
Message
|
| opmind |
 Posted: Thu Jul 19, 2012 4:56 am Post subject: Migrating from HACMP to Multi Instance |
 |
|
Acolyte
Joined: 04 Dec 2001
Posts: 68
|
Hello there.
We plan to remove HACMP for our aix boxes next year and I need to prepare everything to be very successfull.
Actually our Production MQ Network is designed using MQSeries Cluster. To be in a high availability mode, we installed Full repo in two boxes using HACMP (configured to swich only if something hardware goes wrong) in a Active / Standby. MQ is install such as a stand alone MQ but /var/mqm/log & /var/mqm/qmgrs/xxxx + /var/mqm are on a SAN File System and this is a HACMP Process to be sure everything is switching well if it is needed. After that, MQ is restarting and everything runs like this since 10 years.
I actually read every IBM pdf to be sure to take the best choice concerning our new infra, which is supposed to be without HACMP of course.
I saw something with MQ Multi Instance but it is really not clear ... especially concerning this configure option on an already exist Queue Manager, plus on an already exist Queue Manager full Cluster Repository. As everyone know, it is really touchy to modify something in a Cluster Environment and I really don't want to loose everything because of this change.
Question, is it possible to migrate as I described ?
Does anyone already done this ?
Thanks a lot. |
|
| Back to top |
|
 |
| exerk |
| Posted: Thu Jul 19, 2012 6:20 am Post subject: |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
From the V7.1 Info Centre:
| Quote: |
| ...As discussed previously in this topic, as long as two full repositories are configured, they are almost by their nature highly available. If you need to, WebSphere MQ software High Availability / Multi-instance queue managers can be used for full repositories. There is no strong reason to use these methods, and in fact for temporary outages these methods might cause additional performance cost during the failover... |
Personally, if the FRs are just that and do no application-related work, I'd move the FRs to another couple of queue managers created for the purpose and decommission the old one along with the HACMP.
_________________
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 |
|
|
| opmind |
| Posted: Thu Jul 19, 2012 11:22 pm Post subject: |
|
|
Acolyte
Joined: 04 Dec 2001
Posts: 68
|
There is Application on my Queue manager, not only Cluster Full Repo.
To give more Information, here's a good start with this schema of our Architecture MQSeries.
Because we don't actually do HA on Applications (only on Hardware) is it necessary to implement MQ Instance ? |
|
| Back to top |
|
|
| exerk |
| Posted: Fri Jul 20, 2012 12:06 am Post subject: |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
| opmind wrote: |
| Because we don't actually do HA on Applications (only on Hardware) is it necessary to implement MQ Instance ? |
So, let me get this straight - in the event of failure of the server, the queue manager and Broker fail over? And by 'Applications' you mean these are static on the server and do not fail over?
_________________
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 |
|
|
| opmind |
| Posted: Fri Jul 20, 2012 12:20 am Post subject: |
|
|
Acolyte
Joined: 04 Dec 2001
Posts: 68
|
Yes, if the Server is failing over (which append two times in 10 years), everything is restarted on the BOX 2 (in less than 2 mn ... minutes that we accept ).
Oups, Applications are Message flow on this schema ... but I have two another HA cluster named MQFactory and thoose ones are defined to manage Mq Connexions on the cluster PROD (by MQ Client through WAS & JBOSS).
In thoose HA boxes too, it's only hardware and Active / Standby HA. |
|
| Back to top |
|
|
| exerk |
| Posted: Fri Jul 20, 2012 12:27 am Post subject: |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
| opmind wrote: |
| I have two another HA cluster named MQFactory and thoose ones are defined to manage Mq Connexions on the cluster PROD (by MQ Client through WAS & JBOSS). |
I would expect WAS to MI aware (depending on WAS version) but I don't know about JBOSS. From what you've posted, with an average of one fail-over every five years, and the fact that fail-over was hardware-only (which suggests that the middleware infrastructure is not viewed as critical), do you really need the complexity of high-availability?
_________________
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
|
|
|
|
