| Author |
Message
|
| jeevan |
 Posted: Mon Jul 26, 2010 5:41 am Post subject: Technologies for MQ DR (HACMP/VCS/MI) |
 |
|
Grand Master
Joined: 12 Nov 2005
Posts: 1432
|
We are working on preparing a DR plan for MQ and I am given the resp to come up how to do we do it technically.
The DR site is not in the same datacentre as the current( being in the same location does not make sense) to make a DR site.
We are using MQ 7.0.1.1 on Unix/Linux/indows and our DR wiwll focus on the MQ on Unix and Linux. We can not use the MI features as GPFS does not work in geographically spread area. We are doing some tier 1 SAN replication, but I am not sure weather HACMP or VCS work in case of replicated SAN. Could some one give me some idea who have implemented HACMP for MQ DR?
Do any one know does IBM still supporting the HACMP for MQ 7.x?
Thanks |
|
| Back to top |
|
 |
| Vitor |
| Posted: Mon Jul 26, 2010 6:00 am Post subject: Re: Technologies for MQ DR (HACMP/VCS/MI) |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| jeevan wrote: |
| Could some one give me some idea who have implemented HACMP for MQ DR? |
I've never seen it used for that & don't believe you can (as you intimate). HACMP et al are HA products; DR tends to be a different kettle of fish.
| jeevan wrote: |
| Do any one know does IBM still supporting the HACMP for MQ 7.x? |
AFAIK yes. What leads you to the fear they are not?
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| exerk |
| Posted: Mon Jul 26, 2010 6:05 am Post subject: Re: Technologies for MQ DR (HACMP/VCS/MI) |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
| Vitor wrote: |
| ...What leads you to the fear they are not?... |
Because the MC91 SupportPac has been withdrawn, possibly?
_________________
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 |
|
|
| mqjeff |
| Posted: Mon Jul 26, 2010 6:13 am Post subject: Re: Technologies for MQ DR (HACMP/VCS/MI) |
|
|
Grand Master
Joined: 25 Jun 2008
Posts: 17447
|
| exerk wrote: |
| Vitor wrote: |
| ...What leads you to the fear they are not?... |
Because the MC91 SupportPac has been withdrawn, possibly? |
Alas, MC91 was always Category 2. This means "IBM Supplied but NOT supported by IBM".
HACMP has *never* been "supported". |
|
| Back to top |
|
|
| exerk |
| Posted: Mon Jul 26, 2010 6:19 am Post subject: Re: Technologies for MQ DR (HACMP/VCS/MI) |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
| mqjeff wrote: |
| exerk wrote: |
| Vitor wrote: |
| ...What leads you to the fear they are not?... |
Because the MC91 SupportPac has been withdrawn, possibly? |
Alas, MC91 was always Category 2. This means "IBM Supplied but NOT supported by IBM".
HACMP has *never* been "supported". |
 I stand (humbly) corrected...
_________________
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 |
|
|
| bruce2359 |
| Posted: Mon Jul 26, 2010 6:32 am Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9469
Location: US: west coast, almost. Otherwise, enroute.
|
| Quote: |
| DR tends to be a different kettle of fish. |
Omitting the fish reference, what exactly are the defined objectives of your DR plan?
DR plans are very different from operational 2nd site plans. The vision statement I helped a client develop for their DR plan was "In a disaster, to survive, all restrictions are removed."
In a true disaster (life-death of the organization), this means that the usual operational rules no longer apply. In the first few days or weeks, HA no longer becomes a requirement. In a true disaster, any suitable platform or configuration will suffice.
What exactly is to be recovered? In what time-frame? If everything is to be replicated, then buy a 2nd instance of all hardware and software; and deploy it further away from your data center.
If only one or a few apps are to be recovered during the disaster, then you have a smaller and more manageable task.
_________________
I like deadlines. I like to wave as they pass by.
ב''ה
Lex Orandi, Lex Credendi, Lex Vivendi. As we Worship, So we Believe, So we Live. |
|
| Back to top |
|
|
| jeevan |
| Posted: Mon Jul 26, 2010 6:39 am Post subject: Re: Technologies for MQ DR (HACMP/VCS/MI) |
|
|
Grand Master
Joined: 12 Nov 2005
Posts: 1432
|
| exerk wrote: |
| Vitor wrote: |
| ...What leads you to the fear they are not?... |
Because the MC91 SupportPac has been withdrawn, possibly? |
You are right. But there are a few more reasons. MQ7 has MI features which is an equvalent to HACMP. In that context, and most proabaly IMB may want to promote the MI technology, over the old HACMP. Stick and carrot approach. One can make people compelled with a soft approaches( like withdrawing the support for the tech which does not like to be survived) wihtout/instead of telling people not to use it.
Last edited by jeevan on Mon Jul 26, 2010 6:53 pm; edited 1 time in total |
|
| Back to top |
|
|
| exerk |
| Posted: Mon Jul 26, 2010 6:46 am Post subject: Re: Technologies for MQ DR (HACMP/VCS/MI) |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
| jeevan wrote: |
| ...MQ7 has MI features which is an equivalent to HACMP. In that context, and most probably IMB may want to promote the MI technology, over the old HACMP... |
Being aware of course of limitations in an heterogenous WMQ environment?
_________________
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 Jul 26, 2010 6:53 am Post subject: Re: Technologies for MQ DR (HACMP/VCS/MI) |
|
|
Grand Master
Joined: 24 Feb 2009
Posts: 1343
Location: Floating in space on a round rock.
|
| Vitor wrote: |
| jeevan wrote: |
| Could some one give me some idea who have implemented HACMP for MQ DR? |
I've never seen it used for that & don't believe you can (as you intimate). HACMP et al are HA products; DR tends to be a different kettle of fish. |
Agree !
I believe MQ would be as an application for HACMP Setup.
And for applications, generally, we need to check following start and stop scripts : (in 'How to define an application to HACMP ?')
* Environment is what is expected ( verify prereq conditions satisfied ? that include access to FS, IP Labels and free FS space.)
* Multiple Instances Issues ( when starting an appl. with multiple instances, only start the instances applicable for each node )
* Location of the scripts ( scripts should be available and executable on all nodes of the resource group)
* Handle errors from previous termination (Was previous termination successful ? Is data recovery needed ?)
* Correct coding ( not sure, would it be applicable here for mq ).
_________________
*Life will beat you down, you need to decide to fight back or leave it. |
|
| Back to top |
|
|
| Vitor |
| Posted: Mon Jul 26, 2010 7:19 am Post subject: Re: Technologies for MQ DR (HACMP/VCS/MI) |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| jeevan wrote: |
| MQ7 has MI features which is an equvalent to HACMP. In that context, and most proabaly IMB may want to promote the MI technology, over the old HACMP. |
With respect no. AFAIK the v7 MI has never been promoted as a replacement to the kind of failover software that HACMP represents but an additional set of features that simply the task to those sites which don't have HACMP or simlar. I also feel certain that given the choice between selling a WMQv7 license or selling a WMQv7 license and an HACMP license the average IBM (or IMB) sales rep would have a clear preference.
You can't do everything with MI that you can with HACMP. This is going to remain a fact for some time to come. As is the fact you can't do DR with HACMP.
_________________
Honesty is the best policy.
Insanity is the best defence.
Last edited by Vitor on Mon Jul 26, 2010 7:39 am; edited 1 time in total |
|
| Back to top |
|
|
| bruce2359 |
| Posted: Mon Jul 26, 2010 7:28 am Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9469
Location: US: west coast, almost. Otherwise, enroute.
|
This, like many other OPs, appears to start with a solution in mind - we want to do DR with HACMP and v7... or whatever.
I'd suggest asking management for some definition. For example:
"What constitutes a disaster?"
"What apps must be recovered, and in what time-frame?"
"Is HACMP (or any other technology solution) mandatory in a disaster?"
_________________
I like deadlines. I like to wave as they pass by.
ב''ה
Lex Orandi, Lex Credendi, Lex Vivendi. As we Worship, So we Believe, So we Live. |
|
| Back to top |
|
|
| jeevan |
| Posted: Mon Jul 26, 2010 8:13 am Post subject: |
|
|
Grand Master
Joined: 12 Nov 2005
Posts: 1432
|
| bruce2359 wrote: |
This, like many other OPs, appears to start with a solution in mind - we want to do DR with HACMP and v7... or whatever.
I'd suggest asking management for some definition. For example:
"What constitutes a disaster?"
"What apps must be recovered, and in what time-frame?"
"Is HACMP (or any other technology solution) mandatory in a disaster?" |
There is not whole lot of clarity in management. As you guys already know when they konw this has to be done, they just want to be done without being very clear why we are doing it for. However, at this juncture of time, we want to plan a DR in case of our primary datacentre fails. So, we are not planning for a catastrophic DR, but datacentre falure. Clear in a way but not very much comprehensive in other way. To me, there is nothing wrong to start with something.
That said, my vision to this post is to figure out/ explore the technologies to deploy to make a DR plan in the above context.
I can guess but want to be sure that HACMP works in geographically spread areas. My unix admin said that GPFS only works locally. |
|
| Back to top |
|
|
| Vitor |
| Posted: Mon Jul 26, 2010 8:22 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| jeevan wrote: |
| However, at this juncture of time, we want to plan a DR in case of our primary datacentre fails. So, we are not planning for a catastrophic DR, but datacentre falure. |
So what would you consider a catasophie, if the whole datacenter going out is just an inconvenience?? 
| jeevan wrote: |
| I can guess but want to be sure that HACMP works in geographically spread areas. My unix admin said that GPFS only works locally. |
My earlier comments apply. Given I have an apparently different definition of "disaster" to yours. Lower threashold of panic perhaps due to my age.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| exerk |
| Posted: Mon Jul 26, 2010 8:22 am Post subject: |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
1. Duplicate the infrastructure.
2. Shadow copy to that infrastructure.
3. Await the great 'switch-on'.
Throw that pebble in the management pond and see what ripples are caused - especially when they ask the question about cost 
_________________
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 Jul 26, 2010 8:28 am Post subject: |
|
|
Grand Master
Joined: 24 Feb 2009
Posts: 1343
Location: Floating in space on a round rock.
|
| jeevan wrote: |
| I can guess but want to be sure that HACMP works in geographically spread areas. |
Logically It should (if one is not sure about it) .
If you take a scenario of a network which has 2 nodes ( say servers) with a Service IP ( i.e. is a part of Resource Group) which can move between nodes ( if Node1 fails it would move to Node2 ) on failover. {Service IPs are configured over non-service IPs i.e. Boot IPs of nodes}.
Based on the Nature and functionality, Service IP is given to the applications (like MQ/Oracle/DB2) for configuration.
So, Nodes can be geographically spread over the network.
Edit : It also depends on the network topology being used !!!! BUT that is a concern for network/storage while framing it up.
_________________
*Life will beat you down, you need to decide to fight back or leave it.
Last edited by shashivarungupta on Mon Jul 26, 2010 8:38 am; edited 1 time in total |
|
| Back to top |
|
|
|
|
