| |
|
RSS Feed - WebSphere MQ Support
|
RSS Feed - Message Broker Support
|
  |
|
| Fixpack installation: Prod before Dev? |
« View previous topic :: View next topic » |
| Author |
Message
|
| SAFraser |
 Posted: Wed Feb 20, 2008 12:18 pm Post subject: Fixpack installation: Prod before Dev? |
 |
|
Shaman
Joined: 22 Oct 2003
Posts: 742
Location: Austin, Texas, USA
|
What is your practice in the following situation?
You have slogged through all your development servers and most of your production servers, upgrading to v6.x and applying the latest FP (say, #4).
Just as you are ready to upgrade the last of your production servers, FP 5 is issued.
What do you install on these last production servers: FP4 or FP5?
Thanks,
Shirley |
|
| Back to top |
|
 |
| jefflowrey |
| Posted: Wed Feb 20, 2008 12:27 pm Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
If I were a change manager, I'd deploy a large trout to any Admin that put FP5 down, when only FP4 had been tested.
There is at least some predictability for when FPs will be released. Change cycles can be adjusted to be better in tune to prevent this.
If there are critical fixes in the FP, you can maybe skip putting it on Dev, and only rerun it through Perf/QA (whatever the last stage before Prod is). If it fails testing there, then you have to defer it to the next change cycle. If it doesn't, then it depends on local change management procedures, I'd say.
The situation is entirely different, mind you, if the FP4 has been *withdrawn* and *replaced* with FP5.
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Wed Feb 20, 2008 4:56 pm Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7722
|
Since every new release introduces new bugs you must roll out FP5 thru the test environments first to make sure none of the new bugs effects you.
| jefflowrey wrote: |
| The situation is entirely different, mind you, if the FP4 has been *withdrawn* and *replaced* with FP5. |
Even here I would still roll FP5 out to the test environments first, but in this case you do in much faster to get FP4 overlaid with FP5 in Production that much sooner.
Slapping FP5 on in Production before testing could make a bad situation worse regardless of why FP4 is being pulled.
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
| SAFraser |
| Posted: Wed Feb 20, 2008 5:33 pm Post subject: |
|
|
Shaman
Joined: 22 Oct 2003
Posts: 742
Location: Austin, Texas, USA
|
Since the first two responses argue against FP5 to production, I will be devil's advocate and add this to the mix.
From a practical perspective, considering all the CSDs/FPs over the years for the various versions of MQ, how many introduced a significant bug with fairly universal impact? (I can think of only one.)
From a practical perspective, if a maintenance window requires 3 months to negotiate and another month or two to execute, which is riskier? Applying FP5 too soon or applying it too late? Is it more likely to fix a lot of things while unlikely to break much of anything? Or not?
You may think I am arguing in favor of FP5 straight to production. I am not. I have purposely not expressed my own opinion because I wanted a discussion. |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Wed Feb 20, 2008 7:38 pm Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7722
|
By the time I rolled 6.0.2.1 to my last production box I had to apply 4 hot fixes, 2 of which I discovered. Each time I went back to LAB then DEV and finally QA.
I was not worried that 6.0.2.2 was released before I got 6.0.2.1 everywhere. If you played that game you would never win.
| Quote: |
| From a practical perspective, considering all the CSDs/FPs over the years for the various versions of MQ, how many introduced a significant bug with fairly universal impact? (I can think of only one.) |
Using that logic, why bother rolling anything out to DEV and QA first? 
_________________
Peter Potkay
Keep Calm and MQ On |
|
| Back to top |
|
|
| jsware |
| Posted: Thu Feb 21, 2008 12:23 am Post subject: |
|
|
Chevalier
Joined: 17 May 2001
Posts: 455
|
| SAFraser wrote: |
| From a practical perspective, considering all the CSDs/FPs over the years for the various versions of MQ, how many introduced a significant bug with fairly universal impact? (I can think of only one.) |
MQv5.3 CSDs 8/9/10/11 are ones I would avoid because of Java problems well documented on this forum. I remember MQ techies getting grilled on CSD 3 at a conference (a significant memory leak I think but I forget the MQ version - possibly 5.2).
_________________
Regards
John
The pain of low quaility far outlasts the joy of low price. |
|
| Back to top |
|
|
| Vitor |
| Posted: Thu Feb 21, 2008 2:35 am Post subject: |
|
|
Grand High Poobah
Joined: 11 Nov 2005
Posts: 26093
Location: Texas, USA
|
| SAFraser wrote: |
| From a practical perspective, considering all the CSDs/FPs over the years for the various versions of MQ, how many introduced a significant bug with fairly universal impact? (I can think of only one.) |
From a management risk perspective you'd only need one.
You also don't need a bug with a universal impact. Consider the case where the new maintenance contains a bug where message grouping is not handled corrected (say last message in group is not flagged to the receiving program in all instances). It's hardly a universal impact as any site using request/reply or datagrams will be unaffected. But if your site relies on message groups, or perhaps uses message groups in only one application that's business critical, you're in a world of hurt.
_________________
Honesty is the best policy.
Insanity is the best defence. |
|
| 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
|
|
|
|
