|
RSS Feed - WebSphere MQ Support
|
RSS Feed - Message Broker Support
|
 |
|
error for global coordinated message flow |
« View previous topic :: View next topic » |
Author |
Message
|
lium |
Posted: Wed Jul 17, 2002 6:41 am Post subject: error for global coordinated message flow |
|
|
Disciple
Joined: 17 Jul 2002 Posts: 184
|
Hi,guys:
I have a message flow which will update the db2 database in the compute node.I want to configure it as global coordinated.
The MQSI message flow works pretty well if it isn't global. But once I specified it as "Coordinated Transaction"
during deployment, then It never works. The message will be put into backoutq or dead letter queue.
I turned on the tracing, message from it looks like:
2002-07-16 17:42:57.019001 2084 ImbOdbcHandle::checkRcInner , 51022, -868, [IBM][CLI Driver] SQL0868N A CONNECT using a USER/USING clause was attempted to a server for which a connection already exists. SQLSTATE=51022
At the same time, there is error log in the db2diag.log. There are ncreaing items like:
2002-07-16-16.59.43.825000 Instance:DB2 Node:000
PID:2140(DataFlowEngine.exe) TID:2240 Appid:*LOCAL.DB2.020716204449
XA DTP Support sqlxa_open Probe:7
I also tried to use the database node. The message sit in the input queue seems like nothing happened. I also turned on the tracing.
No record for this message. It seems that the message never be processed. It just remains in the input queue.
Then I wrote a simplest db operation, just one db update. It is the same. I tested it. It isn't global coordinated, everthing is fine.
But global coordinated, the same as previous message flow.
Normally, I will think some configuration is wrong. But The queue manager has been configured to be xa manager. I verified it by the sample program which is located
under c:\program files\MQSeries\tools\c\samples\xatm. These two program amqsxag0 and amqsxas0 works pretty well.
The db operation and message retrieving are within UOW. It seems that no problem both queue manager and db2
have been configured in right way.
What is more, I found when a user exception is thrown, the previouse db operation will never roll back.
It conflicts with what it says in the MQSI doc. (This is the reason why I want to configure message flow as global coordinated)
Could you tell me is that true or if you also face the same situation.
Appreciated for any help.
Michael |
|
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
|
|
|
|