Author |
Message
|
KeithPark |
Posted: Thu Jun 09, 2005 6:13 am Post subject: |
|
|
 Apprentice
Joined: 12 Nov 2002 Posts: 25
|
Hi
Latest update on this problem is that the corruption is size related.
When a the messages being posted are less than 4Kb then they are read correctly, however, once the message size creeps above that magic number then the second starts to corrupt the first.
When it was running correctly it was recovering the messages in the correct order.
Keith |
|
Back to top |
|
 |
jefflowrey |
Posted: Thu Jun 09, 2005 7:37 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002 Posts: 19981
|
4Kb, or 4Mb?
My test code works fine for me when I run against 14kb messages.
****************
EDIT
****************
No. It doesn't!
It shows the exact same behavior... (It always helps to test with *different* data ) _________________ I am *not* the model of the modern major general. |
|
Back to top |
|
 |
fjb_saper |
Posted: Thu Jun 09, 2005 12:11 pm Post subject: |
|
|
 Grand High Poobah
Joined: 18 Nov 2003 Posts: 20756 Location: LI,NY
|
Never seen this type of behavior. But then I am nearly always using JMS... |
|
Back to top |
|
 |
KeithPark |
Posted: Mon Jun 13, 2005 6:54 am Post subject: |
|
|
 Apprentice
Joined: 12 Nov 2002 Posts: 25
|
Latest breaking news...
A work around solution has been found.
When the message is read from the queue I am now reading the data into a byte array which I then use to populate a second message which is placed into the array. This is messy, but provides the decoupling which appears to be required.
Code: |
int msgLen = theMessage.getMessageLength () ;
System.out.println("Message Length = " + msgLen);
byte [] msgData = new byte [msgLen] ;
theMessage.readFully (msgData) ;
MQMessage mqMsg = new MQMessage();
mqMsg.write(msgData,0,msgLen);
vMessages.addElement((Object)mqMsg);
|
 |
|
Back to top |
|
 |
jefflowrey |
Posted: Mon Jun 13, 2005 6:58 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002 Posts: 19981
|
Did you open a PMR with IBM? _________________ I am *not* the model of the modern major general. |
|
Back to top |
|
 |
KeithPark |
Posted: Mon Jun 13, 2005 7:02 am Post subject: |
|
|
 Apprentice
Joined: 12 Nov 2002 Posts: 25
|
Yep.
PMR is open & still chasing them to provide a real solution.
 |
|
Back to top |
|
 |
KeithPark |
Posted: Thu Jun 16, 2005 4:52 am Post subject: |
|
|
 Apprentice
Joined: 12 Nov 2002 Posts: 25
|
Latest update...
IBM have agreed that this is a problem with the CSD 9 jars and have given me a fix to test.
 |
|
Back to top |
|
 |
jefflowrey |
Posted: Thu Jun 16, 2005 4:55 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002 Posts: 19981
|
KeithPark wrote: |
Latest update...
IBM have agreed that this is a problem with the CSD 9 jars and have given me a fix to test.
 |
Tell them it is also a problem with the v6 Beta jars, as well. _________________ I am *not* the model of the modern major general. |
|
Back to top |
|
 |
PeterPotkay |
Posted: Sat Jun 18, 2005 10:06 am Post subject: |
|
|
 Poobah
Joined: 15 May 2001 Posts: 7722
|
|
Back to top |
|
 |
jefflowrey |
Posted: Sat Jun 18, 2005 5:55 pm Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002 Posts: 19981
|
Yes.
I would have hoped that such things would not have been in the v6 beta. _________________ I am *not* the model of the modern major general. |
|
Back to top |
|
 |
KeithPark |
Posted: Wed Jun 29, 2005 12:35 am Post subject: |
|
|
 Apprentice
Joined: 12 Nov 2002 Posts: 25
|
Quote: |
I would have hoped that such things would not have been in the v6 beta. |
I have notified the chaps who provided my fix that they need to take a look at the V6 beta release as well.
 |
|
Back to top |
|
 |
|