Author |
Message
|
shalabh1976 |
Posted: Fri Oct 19, 2007 5:13 am Post subject: Better way of error/exception handling |
|
|
 Partisan
Joined: 18 Jul 2002 Posts: 381 Location: Gurgaon, India
|
The following link gives a standard approach to exception handling:
http://www.ibm.com/developerworks/websphere/library/techarticles/0305_ansari/ansari.html
However I see existing flows in my current project having all the nodes with their failure terminals wired to a trace node. The current logic is to capture the state of the input message in the node in the case of an exception. I know that if the standard error handling mechanism is used and I capture the ExceptionList I still cannot get the state of the input message in the exception causing node in my exception handler.
However in all my years of MB development I have not needed to look at the state of the input message explicitly and the exceptionlist tree has been good enough.
Is there a good reason why I should not change the existing design to one suggested by Thameem.
After all I am not the one who will support the application after development. _________________ Shalabh
IBM Cert. WMB V6.0
IBM Cert. MQ V5.3 App. Prog.
IBM Cert. DB2 9 DB Associate |
|
Back to top |
|
 |
jefflowrey |
Posted: Fri Oct 19, 2007 5:38 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002 Posts: 19981
|
You notice that that article is for v2.1, right?
You should look for an article (or maybe it's a redpiece?) called "Bullet-proofing message flows" or similar. _________________ I am *not* the model of the modern major general. |
|
Back to top |
|
 |
dkeister |
Posted: Fri Oct 19, 2007 5:41 am Post subject: |
|
|
Disciple
Joined: 25 Mar 2002 Posts: 184 Location: Purchase, New York
|
|
Back to top |
|
 |
shalabh1976 |
Posted: Fri Oct 19, 2007 5:42 am Post subject: |
|
|
 Partisan
Joined: 18 Jul 2002 Posts: 381 Location: Gurgaon, India
|
Jeff,
It's a redbook. For V6. Downloading it now.
However the client is on v5.
Will check how much from the redbook is applicable.
Good reading for the weekend. _________________ Shalabh
IBM Cert. WMB V6.0
IBM Cert. MQ V5.3 App. Prog.
IBM Cert. DB2 9 DB Associate |
|
Back to top |
|
 |
jefflowrey |
Posted: Fri Oct 19, 2007 5:48 am Post subject: |
|
|
Grand Poobah
Joined: 16 Oct 2002 Posts: 19981
|
Most of it is directly applicable, I believe. _________________ I am *not* the model of the modern major general. |
|
Back to top |
|
 |
dkeister |
Posted: Fri Oct 19, 2007 6:23 am Post subject: |
|
|
Disciple
Joined: 25 Mar 2002 Posts: 184 Location: Purchase, New York
|
Yes, the information is applicable. I developed the material on V5. The principles and details remain the same for V6. _________________ Dean Keister |
|
Back to top |
|
 |
shalabh1976 |
Posted: Fri Oct 19, 2007 8:45 am Post subject: |
|
|
 Partisan
Joined: 18 Jul 2002 Posts: 381 Location: Gurgaon, India
|
Dean,
Please refer to page 17.
The following statement:
set msgbody = r.LFROW.ROW;
makes no sense.
The value of r.LFROW.ROW is not set anywhere.
Am I missing something? _________________ Shalabh
IBM Cert. WMB V6.0
IBM Cert. MQ V5.3 App. Prog.
IBM Cert. DB2 9 DB Associate |
|
Back to top |
|
 |
dkeister |
Posted: Fri Oct 19, 2007 9:23 am Post subject: |
|
|
Disciple
Joined: 25 Mar 2002 Posts: 184 Location: Purchase, New York
|
Sorry I was not clear. This was only a portion of the code. There is a statement:
Declare r reference to InputBody;
the message set has the structure.
LFROW.ROW
So
set msgbody = r.LFROW.ROW;
picks up the input message. Also, this code assumes the input message is XML and it is creating lines in the output where the XML tags are not split.
If you want more detail, I can send you the full esql, msgset, etc. (dean@mqexerciser.com) _________________ Dean Keister |
|
Back to top |
|
 |
|