| |
|
RSS Feed - WebSphere MQ Support
|
RSS Feed - Message Broker Support
|
  |
|
| fmcojagt > fmcjapi issues |
« View previous topic :: View next topic » |
| Author |
Message
|
| bml137 |
 Posted: Thu Jul 20, 2006 10:04 am Post subject: fmcojagt > fmcjapi issues |
 |
|
Newbie
Joined: 20 Jul 2006
Posts: 2
|
OS: AIX
We are in the process of moving our applications from v3.4 to v3.6. I am in a development/test environment right now and trying to see what issues we run into. I tested just about all the code we are using with MQWF and have run into 2 issues:
(1) When I run the exact same code with 'fmcojagt.jar.deprecated', it works. When I run it with 'fmcjapi.jar', the 'ProcessInstance.start2' method causes all 'fmcemain' processes to crash. The IBM Web Client can start workitems, everything else works (workitems advance through the workflow), but start2 crashes. If I comment just that line out, everything runs fine (except the item isn't started, of course).
(2) Using fmcjapi.jar (I haven't tried the old jar yet), 'Agent.locate' returns 'null'. It does not throw an exception...it just returns 'null'. I tried passing in the system-group/system and I also tried empty strings. I tried setting the configuration ID and using LOC_LOCATOR. Essentially, it can't find workflow. Here's the kicker. This is only the case when running from the shell. When I do an 'Agent.locate' from code running in WebSphere, it returns an ExecutionService.
Thanks,
Brian |
|
| Back to top |
|
 |
| bml137 |
| Posted: Thu Jul 20, 2006 11:46 am Post subject: |
|
|
Newbie
Joined: 20 Jul 2006
Posts: 2
|
As an addition to (2):
I switched to the 'fmcojagt.jar' library and it works now. It appears that I cannot get either the 'Agent.locate' or 'ProcessInstance.start2' methods to work with the 'fmcjapi.jar' library.
fyi: When trying to get 'Agent.locate' to work, I even added 'fmcrc' to the classpath. |
|
| Back to top |
|
|
| hos |
| Posted: Thu Jul 20, 2006 10:58 pm Post subject: |
|
|
Chevalier
Joined: 03 Feb 2002
Posts: 470
|
Well,
that's by no way enough information to make a serious assessment but my first guess would be that your plain (shell) applications do not have access to the profile.
Is the AIX box where you are running the applications the same where you have configured the clients? In WebSphere your applications have two ways to find the profile information: either via the fmcrc file hierarchy created during the configuration or via the fmcrc file buried in the ear file. Plain applications only have the file hierarchy.
Setting the Classpath to enable profile access is NOT the right way to solve this problem. Check the 3.6 Installation Guide for the Instant Trace facility that was designed to solve such problems . |
|
| 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
|
|
|
|
