| |
|
RSS Feed - WebSphere MQ Support
|
RSS Feed - Message Broker Support
|
  |
|
| ANONYMOUS in message sets in WBI MB 5.0 |
« View previous topic :: View next topic » |
| Author |
Message
|
| Boolean |
 Posted: Thu Nov 06, 2003 12:22 am Post subject: ANONYMOUS in message sets in WBI MB 5.0 |
 |
|
Acolyte
Joined: 05 Jun 2003
Posts: 56
Location: Sweden
|
Does anyone have a good explanation of the ANONYMOUS "grouping field" that appears when message sets are migrated from 2.1 to 5.0?
Most of the message sets runs ok when migrated, but we have problems with "100% CPU" taken when running one of the migrated ones.
/Anders (Boolean) |
|
| Back to top |
|
 |
| kimbert |
| Posted: Thu Nov 06, 2003 4:14 am Post subject: |
|
|
Jedi Council
Joined: 29 Jul 2003
Posts: 5542
Location: Southampton
|
The new message model supports the use of local definitions. This means that elements, simple types, complex type and groups can all be declared as local to their parent.
The default behaviour of 'mqsimigratemsgsets' is to create local definitions for any object which is only used in one place. Most likely, in your migrated message model, the **ANONYMOUS** refers to a local complex type. In your 2.1 model it would have had a name and an identifier. In the migrated 5.0 model, miqsimigratemsgsets has noticed that it is only used once, so it has created an anonymous local complex type.
Note that because the complex type is local, no other element can ever be based on that type. If that is a problem, you can use the -g' flag which forces mqsimigratemsgsets to create everything global. Don't do that unless you need to though - local definitions reduce the complexity of a message set, and make it easier to understand, even if they look a bit strange at first.
Re: the looping problem, I would need more information before suggesting a cause. Its probably nothing to do with the anonymous stuff, though. |
|
| Back to top |
|
|
| Boolean |
| Posted: Thu Nov 06, 2003 11:44 pm Post subject: |
|
|
Acolyte
Joined: 05 Jun 2003
Posts: 56
Location: Sweden
|
Thank's Kimbert!
I've isolated the problem to a flow with one inputnode and one tracenode.
Have sent it with test message and message set to my local IBM contacts and they got the same result (CPU "hanging"). They will send the case to the IBM support as a possible bug.
I'll get back on this when I hear from IBM.
/Anders (Boolean) |
|
| 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
|
|
|
|
