| Author |
Message
|
| scott9 |
 Posted: Thu Feb 08, 2007 10:07 am Post subject: Question about DST patch requirements with WBI |
 |
|
Acolyte
Joined: 11 Jul 2002
Posts: 62
Location: Sacramento,CA
|
I wanted to verify that my interpretation of the IBM site regarding Message Broker patches for DST is correct. The statement below indicates that the fixpack is required only if your ESQL specifies interactions between the time functions identified; therefore, my interpretation is that no fixpack is required for DST patching if you simply use CURRENT_TIMESTAMP in your ESQL. Anybody disagree with this interpretation?
"Customers affected are those:
who have interactions between the TIMESTAMP, GMTTIMESTAMP, TIME or GMTTIME types within ESQL or Mapping node with TIMESTAMP conversion (as this generates ESQL). You are required to apply the Broker/MRM Runtime fix."
Note: The above reference is truncated, so don't accept that this is the only condition that requires DST patching. It is the only piece where I need clarification.
Reference: http://www-1.ibm.com/support/docview.wss?rs=849&context=SSKM8N&context=SS3GH2&context=SSKMAB&dc=D600&uid=swg21246615&loc=en_US&cs=UTF-8&lang=en
Final note: Sometimes, these discussions take an awkward turn in the direction of people suggesting that you should always install the latest fixpack. March 11 is approaching fast and there are extenuating circumstances that would make it difficult for us to apply fixpacks at this time. Thanks for understanding. |
|
| Back to top |
|
 |
| jefflowrey |
| Posted: Thu Feb 08, 2007 10:10 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
| Quote: |
How do I know if my systems could be affected?
All WebSphere Broker releases will be affected by this DST change. In addition, your Operating Systems may also be affected.
Back to Top
What could happen if I fail to update my systems?
The Broker will suffer unpredictable results if operating in a timezone affect by these DST changes.
Further problems may be observed by applications in the situations described below. |
APPLY THE FIXES.
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
|
| scott9 |
| Posted: Thu Feb 08, 2007 10:20 am Post subject: |
|
|
Acolyte
Joined: 11 Jul 2002
Posts: 62
Location: Sacramento,CA
|
| Quote: |
All WebSphere Broker releases will be affected by this DST change. In addition, your Operating Systems may also be affected.
Back to Top
What could happen if I fail to update my systems?
The Broker will suffer unpredictable results if operating in a timezone affect by these DST changes.
Further problems may be observed by applications in the situations described below. |
It is true that all Websphere Brokers are affected by DST and we are patching the affected JRE on each (we will use JTZU); however, my question was specifically directed towards the requirement to apply the fixpack. That is only required if specific conditions exist and I am just looking for concurrence with my interpretation of the conditions. |
|
| Back to top |
|
|
| jefflowrey |
| Posted: Thu Feb 08, 2007 10:29 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
You're taking what I consider to be a very very narrow interpretation of that message. I believe your interpretation is *too* narrow.
I personally do not consider it worth the risk, myself, even if you are correct.
_________________
I am *not* the model of the modern major general. |
|
| Back to top |
|
|
| scott9 |
| Posted: Thu Feb 08, 2007 12:23 pm Post subject: |
|
|
Acolyte
Joined: 11 Jul 2002
Posts: 62
Location: Sacramento,CA
|
Thanks, jefflowrey, for your feedback. I don't think that I articulated my question very well. I perceive that you are suggesting that we not try to interpret the conditions and just install the fixpack, because it is the safest thing to do. I won't try to argue that point, because the thread would get really long and you would probably end up being right anyway. 
My role is to analyze the requirements and present my findings and suggestions to leadership. I may determine that our code does not meet the conditions that warrant installation of the fixpack, but still suggest that we do it because it is the safe thing to do. At least, I will have presented all the facts and research, so that we can make informed decisions. I don't feel comfortable making the suggestion without proper analysis.
It is my interpretation that specifying CURRENT_TIMESTAMP in ESQL does not meet the requirements to install the DST fixpack. I was really just looking for feedback on whether other people share in this interpretation.
I hope that I have articulated in a way that is not combative (not my style) that I'm not posting this as a question about whether or not to install the fixpack. I'm really just looking for a granular interpretation of one small piece of a larger puzzle. |
|
| Back to top |
|
|
| moorej_gl |
| Posted: Thu Feb 08, 2007 1:53 pm Post subject: |
|
|
Apprentice
Joined: 01 Dec 2003
Posts: 35
Location: Madison, WI
|
| I found it very worthwhile to open a PMR for this. Though we weren't trying to interpret the findings one way or another, there are certainly several paths to choose from to achieve DST compliance... even if there are 'extenuating circumstances'. |
|
| Back to top |
|
|
| bbeardsley |
| Posted: Tue Feb 20, 2007 11:54 am Post subject: |
|
|
Acolyte
Joined: 17 Dec 2001
Posts: 52
Location: Dallas, TX, USA
|
| scott9, If you did indeed open a PMR for this question, I would be interested to hear the response from IBM. |
|
| Back to top |
|
|
| scott9 |
| Posted: Tue Feb 20, 2007 12:03 pm Post subject: |
|
|
Acolyte
Joined: 11 Jul 2002
Posts: 62
Location: Sacramento,CA
|
I did open a PMR. They said that I should install either the csd or APAR fix. I hope this helps. See their reply below:
"...using CURRENT_TIMESTAMP and TIMESTAMP will give a result that is one hour out for two weeks each year until you apply the fix. Whether this matters to you depends on what calculations you make from the time stamps. For example if you use them to determine the local time of day then you will get the wrong answer for a few weeks per year, but if you use them to measure how much time elapsed between two events, then you will get the right answer so long as your time interval does not cover the old clock change." |
|
| Back to top |
|
|
| GEMO! |
| Posted: Tue Mar 27, 2007 4:34 pm Post subject: |
|
|
Apprentice
Joined: 08 Oct 2004
Posts: 33
Location: This World
|
I'm having similar issues.... I'm looking for a resolution.
_________________
GEMO! |
|
| Back to top |
|
|
| scott9 |
| Posted: Tue Mar 27, 2007 4:45 pm Post subject: |
|
|
Acolyte
Joined: 11 Jul 2002
Posts: 62
Location: Sacramento,CA
|
| The resolution is that you must either install FP09 or the APAR fix for WBIV5 or CSD03 for WBIV6. |
|
| Back to top |
|
|
| fjb_saper |
| Posted: Wed Mar 28, 2007 2:02 pm Post subject: |
|
|
Grand High Poobah
Joined: 18 Nov 2003
Posts: 20756
Location: LI,NY
|
| Quote: |
| It is my interpretation that specifying CURRENT_TIMESTAMP in ESQL does not meet the requirements to install the DST fixpack. I was really just looking for feedback on whether other people share in this interpretation. |
I would argue to the opposite.
If you used CURRENT_GMTTIMESTAMP you should be safe.
Enjoy 
_________________
MQ & Broker admin |
|
| Back to top |
|
|
|
|
