| |
|
RSS Feed - WebSphere MQ Support
|
RSS Feed - Message Broker Support
|
  |
|
| Message Size Issue in WMQI |
« View previous topic :: View next topic » |
| Author |
Message
|
| KMcKean |
 Posted: Tue Jun 29, 2004 2:37 am Post subject: Message Size Issue in WMQI |
 |
|
Newbie
Joined: 24 Jul 2003
Posts: 6
|
We are processing large messages consisting of repeating sub messages and whenever we get a message over 50K in size WMQI doesn't seem to be able to cope with the message. It gets to a certain point in the processing of the message and then backs it out with the following error:
Publishing to destination BMTWCFG1:SYSTEM.BROKER.ADMIN.REPLY:4366674d677253756233 for user mqsi.
A publication destination is being added to the list of destinations to BMTWCFG1:SYSTEM.BROKER.ADMIN.REPLY:4366674d677253756233 for user mqsi.
The way we have coded it is that WMQI extracts each individual sub message one after the other, reformats it into XML, and then concatenates the output to the existing output into an Environment variable until all repeats have been reformatted. We have experimented using Local Environments variables instead of Environment variables and it seem to speed up the message processing time but we still get the same error.
Has anyone ever seen this error or do you have any ideas as to how we can resolve this issue. From our findings it seem to be pointing towards a space related issue as opposed to a coding issue, do you know of any temporary storage spaces that we may need to manipulate in order to get the messages to process successfully?
Thanks in advance. |
|
| Back to top |
|
 |
| kirani |
| Posted: Tue Jun 29, 2004 10:39 am Post subject: Re: Message Size Issue in WMQI |
|
|
Jedi Knight
Joined: 05 Sep 2001
Posts: 3779
Location: Torrance, CA, USA
|
| KMcKean wrote: |
...and then backs it out with the following error:
Publishing to destination BMTWCFG1:SYSTEM.BROKER.ADMIN.REPLY:4366674d677253756233 for user mqsi.
A publication destination is being added to the list of destinations to BMTWCFG1:SYSTEM.BROKER.ADMIN.REPLY:4366674d677253756233 for user mqsi.
|
Where is the Error message? Can you explain your message flow, like the nodes in your message flow?
_________________
Kiran
IBM Cert. Solution Designer & System Administrator - WBIMB V5
IBM Cert. Solutions Expert - WMQI
IBM Cert. Specialist - WMQI, MQSeries
IBM Cert. Developer - MQSeries
|
|
| Back to top |
|
|
| JohnMetcalfe |
| Posted: Wed Jun 30, 2004 7:45 am Post subject: |
|
|
Apprentice
Joined: 02 Apr 2004
Posts: 40
Location: Edinburgh, Scotland
|
Strangely enough, today we have hit what looks to be the same problem as yourself.
We are processing 10MB messages, the flow fails - without a error message - on a ESQL line that does nothing more than copy the InputRoot to the OutputRoot. We receive the message:
BIP7082I: Node '': Publishing to destination QIX2:SYSTEM.BROKER.ADMIN.REPLY:4366674d677253756233 for user mqsiuid.
A publication destination is being added to the list of destinations to QIX2:SYSTEM.BROKER.ADMIN.REPLY:4366674
d677253756233 for user mqsiuid.
Looks like yours !
Looking in the system logs we see that the Execution group is encountering a severe error, failing and then restarting. We are running WMQI 2.1 CSD3.
We too suspect some sort of size related issue. We are currently working on trying to find a soln so all I can do is say I'll update you when I've more info. |
|
| Back to top |
|
|
| KMcKean |
| Posted: Thu Jul 01, 2004 6:28 am Post subject: More Information |
|
|
Newbie
Joined: 24 Jul 2003
Posts: 6
|
The flows we are using are quite complex but I will try to give you some more information without over complicating things. The message is read in as a BLOB and then gets routed to a subflow. This evaluates the number of repeating submessages. Each invididual submessage is then extracted using a compute node, stored into an environment variable, validated against a message set and reformatted. Each corresponding output is concatenated to another environment variable until all submessages have been reformatted.
The output message will be shorter than the input message but when the flow reaches the repeat that lies around the 50K mark of the input message, processing within the compute node comes to a sudden halt without any error as such, and the publishing to destination message (as shown above) is output onto the usertrace log. The message then gets rolled back to the input node and tries again. Second time around, when the validation takes place, WMQI recognises the backout count is one and therefore fails the message.
Different messages fail at different points in the flow but always around the 50K mark of the input message. Any messages less than 50K are reformatted successfully. We also see in the system logs a severe error encountered in the execution group. We are running WMQI 2.1 CSD4.
Thankyou for your time,
Karen |
|
| Back to top |
|
|
| jefflowrey |
| Posted: Thu Jul 01, 2004 6:35 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002
Posts: 19981
|
The things I would start to look at are:
1) Make sure the message is really valid.
2) Look at the system memory usage during an incident of failure.
3) Enable userlevel trace to see what's going on when the failure occurs.
_________________
I am *not* the model of the modern major general. |
|
| 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
|
|
|
|
