| Author |
Message
|
| Ross |
 Posted: Wed Jun 16, 2010 4:43 am Post subject: Cluster FR joining a second cluster. |
 |
|
Centurion
Joined: 15 Jun 2005
Posts: 127
Location: Ireland
|
Hi.
I have 2 z/OS qmgrs, both as Full Repositories in a large cluster (CLUS1) with 400 ish midrange qmgrs.
I now want to join one of these z/OS qmgrs to another cluster (CLUS2) as part of a new project.
Is there any negatives or possible issues with this?
CLUS1 is integral to our business so my worry is that something in CLUS2 could affect CLUS1.
Any thoughts?
Thanks,
Ross. |
|
| Back to top |
|
 |
| Vitor |
| Posted: Wed Jun 16, 2010 5:14 am Post subject: Re: Cluster FR joining a second cluster. |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| Ross wrote: |
Is there any negatives or possible issues with this?
|
Do a search in here for "overlapping clusters". Much interesting discussion.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| Ross |
| Posted: Wed Jun 16, 2010 5:17 am Post subject: |
|
|
Centurion
Joined: 15 Jun 2005
Posts: 127
Location: Ireland
|
|
| Back to top |
|
|
| aditya.aggarwal |
| Posted: Wed Jun 16, 2010 11:25 am Post subject: |
|
|
Master
Joined: 13 Jan 2009
Posts: 252
|
ROSS Wrote:
| Quote: |
I have 2 z/OS qmgrs, both as Full Repositories in a large cluster (CLUS1) with 400 ish midrange qmgrs.
I now want to join one of these z/OS qmgrs to another cluster (CLUS2) as part of a new project.
Is there any negatives or possible issues with this?
CLUS1 is integral to our business so my worry is that something in CLUS2 could affect CLUS1. |
Are you planning to have the old z/os qmgr as FR in new CLUS2 also? |
|
| Back to top |
|
|
| bruce2359 |
| Posted: Wed Jun 16, 2010 12:31 pm Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9469
Location: US: west coast, almost. Otherwise, enroute.
|
Hey. Quit it. 
_________________
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 |
|
|
| bruce2359 |
| Posted: Wed Jun 16, 2010 12:34 pm Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9469
Location: US: west coast, almost. Otherwise, enroute.
|
And, no, there are no downsides to a z/OS qmgr participating in multiple clusters. Lots of horsepower and reliability there in z/OS. It is not uncommon to have dozens or more qmgrs running in the same LPAR.
_________________
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 |
|
|
| Vitor |
| Posted: Wed Jun 16, 2010 12:36 pm Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| bruce2359 wrote: |
| And, no, there are no downsides to a z/OS qmgr participating in multiple clusters. |
| bruce2359 wrote: |
| Lots of horsepower and reliability there in z/OS. |
Which make them common (and good) choices as FR queue managers in clusters, even if the cluster is mostly distributed systems.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| Back to top |
|
|
| Ross |
| Posted: Wed Jun 16, 2010 3:20 pm Post subject: |
|
|
Centurion
Joined: 15 Jun 2005
Posts: 127
Location: Ireland
|
Thanks for the info.
The z/OS qmgr won't be a FR or PR in this cluster. Just a lowly member!
I appreciate the benefits of processing power, but I'm not involved in the admin of the cluster. |
|
| Back to top |
|
|
| mqjeff |
| Posted: Wed Jun 16, 2010 3:32 pm Post subject: |
|
|
Grand Master
Joined: 25 Jun 2008
Posts: 17447
|
| Ross wrote: |
| The z/OS qmgr won't be a FR or PR in this cluster. Just a lowly member! |
A lowly member *is* a PR. |
|
| Back to top |
|
|
| bruce2359 |
| Posted: Wed Jun 16, 2010 3:32 pm Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9469
Location: US: west coast, almost. Otherwise, enroute.
|
For qmgrs that are part of a cluster there are only two choices: FR or PR.
_________________
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 |
|
|
| Ross |
| Posted: Thu Jun 17, 2010 2:08 am Post subject: |
|
|
Centurion
Joined: 15 Jun 2005
Posts: 127
Location: Ireland
|
My mistake. as thinking there was a PR setting!
Clustering 101 - Fail! |
|
| Back to top |
|
|
| bruce2359 |
| Posted: Thu Jun 17, 2010 5:30 am Post subject: |
|
|
Poobah
Joined: 05 Jan 2008
Posts: 9469
Location: US: west coast, almost. Otherwise, enroute.
|
There is a qmgr attribute that indicates that this qmgr is to be a Full Repos for a named cluster or a cluster in a namelist.
To be a Partial, all that needs to be done is to define a CLUSSDR and CLUSRCVR channel to an FR
_________________
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 |
|
|
| Ross |
| Posted: Fri Jun 18, 2010 1:40 am Post subject: |
|
|
Centurion
Joined: 15 Jun 2005
Posts: 127
Location: Ireland
|
Thanks Bruce. I'm aware of the above, and have that done.
Facing an issue where I define my CLUSSDR with an IP address, and when I join the cluster, my clussdr definition is overridden by the FR's CLUSRCVR definition, which contains a DNS.
Unfortunately this DNS is not known to my mainframe! I'm trying to get the network guys to resolve this, but is there a method within MQ to fix this?
Thanks. |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Fri Jun 18, 2010 3:40 am Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7722
|
Define the CLUSRCVR with an IP address? Not recomended though. All cluster members should be able to resolve all the cluster memebers's DNS names, and you should avoid IP's whenever possible in channel definitions.
Clusters spanning different networks are hairy and scary beasts.
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
| mqjeff |
| Posted: Fri Jun 18, 2010 4:55 am Post subject: |
|
|
Grand Master
Joined: 25 Jun 2008
Posts: 17447
|
| PeterPotkay wrote: |
| Clusters spanning different networks are hairy and scary beasts. |
I think you mean "Clusters spanning different networks are a BAD idea. That way leads to pain and suffering". |
|
| Back to top |
|
|
|
|
