| Author |
Message
|
| PeterPotkay |
 Posted: Wed Nov 25, 2009 5:54 pm Post subject: |
 |
|
Poobah
Joined: 15 May 2001
Posts: 7723
|
In a new implementation, no hassle. If you need a new Highly Available QM I think Multi Instance technology gets the first look over traditional H.A. solutions. Its easier to say all new clients to this new QM will be MQ 7.0.1 or better.
But if you are talking about upgrading an existing QM up to MQ 7.0.1 in a mature environment it will be quite a while before every connecting QM and every MQ Client is upgraded to MQ 7.0.1. We plan on sticking with MC91 on Unix and MSCS on Windows for at least a couple of years on our existing QMs.
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
 |
| exerk |
| Posted: Wed Nov 25, 2009 10:14 pm Post subject: |
|
|
Jedi Council
Joined: 02 Nov 2006
Posts: 6339
|
| PeterPotkay wrote: |
| But if you are talking about upgrading an existing QM up to MQ 7.0.1 in a mature environment it will be quite a while before every connecting QM and every MQ Client is upgraded to MQ 7.0.1. |
My point exactly! And we're certainly looking to use V7.0.1 as client 'gateways' due to the throttling feature.
_________________
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 |
|
|
| student |
| Posted: Thu Nov 26, 2009 4:29 am Post subject: MQCTL on MQv7.0.1 |
|
|
Apprentice
Joined: 17 Mar 2009
Posts: 31
Location: UK
|
| Its looks like resource crunch issues. Please have a look at available memory. you might need to restart the server to resolve the issues at the moment. But in long term IBM should provide the fix. |
|
| Back to top |
|
|
| George Carey |
| Posted: Tue Dec 01, 2009 5:06 pm Post subject: One serious issue |
|
|
Knight
Joined: 29 Jan 2007
Posts: 500
Location: DC
|
One serious issue with the MQv7.0.1 instance setup
It needs to work !!!
Having serious lockd issues with the NFS mounted contact admin volumes !!!
Currently have a QMGR filesystem that got corrupted some how as it
thinks it is running else where but it is not !!!
Some flag(s) is/are hanging it up !!! Can't delete it(dltmqm) , can't do sh@#$% with it.
dspmq just says QMGR running else where(my a$$).
May have to delete the freaking contact admin volumes and recreate them just to get back to a running QMGR!!!
Need a very seriously upgraded amqiclen for this version of MQ methinks!
_________________
"Truth is ... grasping the virtually unconditioned",
Bernard F. Lonergan S.J.
(from book titled "Insight" subtitled "A Study of Human Understanding") |
|
| Back to top |
|
|
| Vitor |
| Posted: Tue Dec 01, 2009 7:08 pm Post subject: Re: One serious issue |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| George Carey wrote: |
| It needs to work !!! |
I think you might need a little more detail than that on the PMR.... 
Though IMHO you're in PMR world now. I'd reference this thread in the PMR if it was me.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| fjb_saper |
| Posted: Wed Dec 02, 2009 11:06 pm Post subject: |
|
|
Grand High Poobah
Joined: 18 Nov 2003
Posts: 20767
Location: LI,NY
|
Are you sure you are using the right version of NFS?
_________________
MQ & Broker admin |
|
| Back to top |
|
|
| George Carey |
| Posted: Thu Dec 03, 2009 11:22 am Post subject: nfs and nfs4 |
|
|
Knight
Joined: 29 Jan 2007
Posts: 500
Location: DC
|
The Netapp device handles nfs = vers 3
and nfs4
The mount settings plus an idmapd configuration to sync with the Netapp filer has to be setup to make nfs4 work else a user and group squash to user group nobody nobody occurs on the mounted volumes such as /var/mqm. Haven't been able to get around that yet. Have calls into Netapp to see why not working. But is works with nfs vers 3 with the verification lock test IBM ships with MQv7.0.1.
But not using multi-instance functionality of MQv7.0.1 though it is configured so I could. IBM L3 has gotten back and said it is a lock and unlock issue but at this time are pointing to Linux as the likely culprit in not release the lock.
An environment flag has to be set to turn off the new multi-instance semaphore lock usage that prevents multiple servers from accessing same MQ data volumes simultaneously. It is on by default for MQv7.0.1 ... With it off kind of behaves like previous versions of MQ and not concerned with multi-server access to data files.
_________________
"Truth is ... grasping the virtually unconditioned",
Bernard F. Lonergan S.J.
(from book titled "Insight" subtitled "A Study of Human Understanding") |
|
| Back to top |
|
|
| bruce2359 |
| Posted: Thu Dec 03, 2009 11:40 am Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9482
Location: US: west coast, almost. Otherwise, enroute.
|
I seem to recall from the MI rollout webinar that nfs4 was required; but I expect to be corrected shortly.
_________________
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 |
|
|
| mvic |
| Posted: Thu Dec 03, 2009 12:15 pm Post subject: |
|
|
Jedi
Joined: 09 Mar 2004
Posts: 2080
|
|
| Back to top |
|
|
| George Carey |
| Posted: Thu Dec 03, 2009 1:59 pm Post subject: Required or Recommended ? |
|
|
Knight
Joined: 29 Jan 2007
Posts: 500
Location: DC
|
Where did you see nfs4 required ??
I see a paragraph that says this:
| Quote: |
| Because not all the locking protocols of network file systems are robust, and because a file system might be configured for performance rather than data integrity, you must run the amqmfsck command to test whether a network file system will control access to queue manager data and logs correctly. |
I see a 'you must' here ! Not elsewhere !!
This means to me if amqmfsck works so will the MQv7.0.1 Locking !!
The section speaks in a preceding paragraph, to the historical background of earlier versions of NFS as not being sufficiently robust but that NFSv4 now was. That was just a backgrounder and recommendation, not a 'you must'.
_________________
"Truth is ... grasping the virtually unconditioned",
Bernard F. Lonergan S.J.
(from book titled "Insight" subtitled "A Study of Human Understanding") |
|
| Back to top |
|
|
| fjb_saper |
| Posted: Fri Dec 04, 2009 4:43 pm Post subject: |
|
|
Grand High Poobah
Joined: 18 Nov 2003
Posts: 20767
Location: LI,NY
|
Well at the presentation I was at, I understood that it would not be working on nfs3 and you needed nfs4 to make the multi-instance qmgr work. 
_________________
MQ & Broker admin |
|
| Back to top |
|
|
| George Carey |
| Posted: Mon Dec 07, 2009 9:05 am Post subject: try it |
|
|
Knight
Joined: 29 Jan 2007
Posts: 500
Location: DC
|
Set it up on nfs v 3 it works.
If the sample amqfsck program works, so will the sample test programs.
Just not recommended.
_________________
"Truth is ... grasping the virtually unconditioned",
Bernard F. Lonergan S.J.
(from book titled "Insight" subtitled "A Study of Human Understanding") |
|
| Back to top |
|
|
| Michael Dag |
| Posted: Tue Dec 08, 2009 1:40 am Post subject: |
|
|
Jedi Knight
Joined: 13 Jun 2002
Posts: 2607
Location: The Netherlands (Amsterdam)
|
|
| Back to top |
|
|
| George Carey |
| Posted: Tue Dec 08, 2009 12:11 pm Post subject: agreed |
|
|
Knight
Joined: 29 Jan 2007
Posts: 500
Location: DC
|
Agreed it needs a more prominent place if indeed it is required.
But IBM has a conditional/CYA even for nfs4.
They say(paraphrasing): Due to implementation differences on different platforms for nfs for performance or other reasons ... blah.. blah a locking verification test must be used to confirm locking and unlocking functionality ... (wording close to this) ... the amqmfsck program is being referred to here.
So even if you have an nfs4 mounted filesystem it had better also pass the amqmfsck test as well. (which it should of course)
Only confusion factor is, so likely does nfs(vers3) pass the amqmfsck test.(mine did)
Having experieced some serious problems with nfs (vers 3) filesystems and moving to nfs4 after some cross platform issues were resolved. I am hoping that nfs4 will make them stay away else MQv7.0.1 has some serious updates needed.
_________________
"Truth is ... grasping the virtually unconditioned",
Bernard F. Lonergan S.J.
(from book titled "Insight" subtitled "A Study of Human Understanding") |
|
| Back to top |
|
|
| Michael Dag |
| Posted: Wed Dec 09, 2009 12:57 am Post subject: |
|
|
Jedi Knight
Joined: 13 Jun 2002
Posts: 2607
Location: The Netherlands (Amsterdam)
|
George,
may be it should read nfs v4 AND amqmqfsck at all times.
anyway I hope your problems will go away and stay away after moving to nfs v4,
please let us know...
_________________
Michael
MQSystems Facebook page |
|
| Back to top |
|
|
|
|
