| Author |
Message
|
| CHF |
 Posted: Mon May 17, 2004 4:29 am Post subject: [SOLVED]No Operational Activity |
 |
|
Master
Joined: 16 Dec 2003
Posts: 297
|
Hi all,
MY WF is behaving weird. I have a process model in which the first activity is a No Operational (Empty activity). I am working on this model for 3 months, and it was working fine. But today I am doing some kind of testing and I put 100 messages to start new Process instances. Out of 100, only 20 of the Instances moved to next activity.
WF 3.4 on OS/390 - SP4
Does anybody have any idea on why its behaving like that?
Thanks
CHF 
Last edited by CHF on Mon May 17, 2004 8:49 am; edited 1 time in total |
|
| Back to top |
|
 |
| jmac |
| Posted: Mon May 17, 2004 5:00 am Post subject: |
|
|
Jedi Knight
Joined: 27 Jun 2001
Posts: 3081
Location: EmeriCon, LLC
|
What is the state of the instances that did not move?
_________________
John McDonald
RETIRED |
|
| Back to top |
|
|
| CHF |
| Posted: Mon May 17, 2004 5:03 am Post subject: |
|
|
Master
Joined: 16 Dec 2003
Posts: 297
|
Process Instace State is RUNNING
No Op activity State is also RUNNING
CHF |
|
| Back to top |
|
|
| jmac |
| Posted: Mon May 17, 2004 5:08 am Post subject: |
|
|
Jedi Knight
Joined: 27 Jun 2001
Posts: 3081
Location: EmeriCon, LLC
|
This is pretty strange... never seen this before, but then again I don't work on 390.
Any messages in fmcsys, fmcerr, db2diag, or MQ logs?
This may be one for IBM, have you opened a PMR?
_________________
John McDonald
RETIRED |
|
| Back to top |
|
|
| CHF |
| Posted: Mon May 17, 2004 5:49 am Post subject: |
|
|
Master
Joined: 16 Dec 2003
Posts: 297
|
Yes, it is strange 
I have just put one test message now, and its working fine, its moving to the next activity. But those 80 instances are still in Running state, not moving to next activity.
May be I will test with 100 more messages and see how it behaves?
I will post updates.
Thanks for the reply Jmac.
CHF |
|
| Back to top |
|
|
| jmac |
| Posted: Mon May 17, 2004 5:57 am Post subject: |
|
|
Jedi Knight
Joined: 27 Jun 2001
Posts: 3081
Location: EmeriCon, LLC
|
Did you do your original testing with the same version of MQWF?
There was a change to the Noop activity in SP3. However, it does NOT look like this is the problem to me, since your symptom does not match the change that was made.
_________________
John McDonald
RETIRED |
|
| Back to top |
|
|
| CHF |
| Posted: Mon May 17, 2004 6:04 am Post subject: |
|
|
Master
Joined: 16 Dec 2003
Posts: 297
|
I dont think there is any problem with the modeling of NoOp, because its working fine even now when I put one test message.
By the way, what is the change to the NoOp activity in SP3? Where can I get the latest modeling instructions for NoOp activity?
Thanks
CHF |
|
| Back to top |
|
|
| jmac |
| Posted: Mon May 17, 2004 6:28 am Post subject: |
|
|
Jedi Knight
Joined: 27 Jun 2001
Posts: 3081
Location: EmeriCon, LLC
|
| CHF wrote: |
| IBy the way, what is the change to the NoOp activity in SP3? Where can I get the latest modeling instructions for NoOp activity? |
The change had to do with the fact that you must now take one path out of a noop or the activity will go InError.
To access the doc go to the IBM search site and serach on MQ Workflow Empty Activity.
_________________
John McDonald
RETIRED |
|
| Back to top |
|
|
| CHF |
| Posted: Mon May 17, 2004 8:48 am Post subject: |
|
|
Master
Joined: 16 Dec 2003
Posts: 297
|
JMac:
Now the model is working fine. I put 1000 messages and all of them moved to next activity. I am still not sure (don't know) what might have happened this morning.
Anyways, Thanks for sharing your ideas.
CHF |
|
| Back to top |
|
|
| jmac |
| Posted: Mon May 17, 2004 9:00 am Post subject: |
|
|
Jedi Knight
Joined: 27 Jun 2001
Posts: 3081
Location: EmeriCon, LLC
|
Just a thought... anything on your Hold queue?
_________________
John McDonald
RETIRED |
|
| Back to top |
|
|
| CHF |
| Posted: Thu May 20, 2004 6:27 am Post subject: |
|
|
Master
Joined: 16 Dec 2003
Posts: 297
|
I see some messages in the HOLD queue. But I dont have access to look up or replay those messages.  I think may be those messages and this problem are due to some database deadlocks that we are getting while we are doing testing.
Anyways, the model is working fine now, and am happy for that.
Thanks for sharing the ideas.
CHF |
|
| Back to top |
|
|
| jmac |
| Posted: Thu May 20, 2004 6:32 am Post subject: |
|
|
Jedi Knight
Joined: 27 Jun 2001
Posts: 3081
Location: EmeriCon, LLC
|
Messages on your hold queue are indicative of some problem. I would assume that these messages are due to your original issues with MQWF. You probably need to figure out why these messages are there (especially if this is a production system).
_________________
John McDonald
RETIRED |
|
| Back to top |
|
|
| CHF |
| Posted: Thu May 20, 2004 1:14 pm Post subject: |
|
|
Master
Joined: 16 Dec 2003
Posts: 297
|
This is the test environment. So we deleted those messages.
Thanks again
CHF |
|
| Back to top |
|
|
|
|
