Symantec IGA

We have a number of tasks in the Task Persistence DB that are in states other than COMPLETE.

According to CA Support, the "process engine" is constantly looking at the Task Persistence DB for work....

But in our situation, we are seeing (intermittently) tasks that are just "hung" and not being processed.

We review the STATE value of the events themselves, and the STATE value of the tasks... and see no pattern as to why they should not be processed.

One example is:

The events for a particular task show:

State=2 is FAILED
State=128 is COMPLETED_STATE… Completed
State=1024 is INITIAL_STATE… In Progress

and the TASK is STATE=1 which is PRE_STATE

Does anyone have an idea how we can "push" the process engine to move forward and work on these "hung" tasks ???

Jimmy,

  There are many things that can cause tasks to hang. Last I knew you are using jboss. If you are not running into DB issues, you could be running into known jboss HornetQ issues.

  This would require a support case to be open for log review, but you should check for warn and error levels for HQ###### type errors. not being able to process message store updates can cause this also.

   If you find them or not, just open a case and reference this communities post.

Thanks,

Bill

Yes... we do see JBoss HornetQ msgs... but not always at the time of the transaction hang.

We do have an open ticket with CA Support, but it is getting very little traction....

08:22:50,092 ERROR [ims.tmt.persistence] (Thread-3670 (HornetQ-client-global-threads-1705684488)) PersistenceProvider: JMS:ID:155ad5f9-db02-11e5-ba63-03544f768a83: fetchObjectById: tried '3' times, exception getting object id: The connection is closed.: com.microsoft.sqlserver.jdbc.SQLServerException: The connection is closed.

at com.microsoft.sqlserver.jdbc.SQLServerException.makeFromDriverError(SQLServerException.java:190) [sqljdbc4.jar:]

We also restarted the JBoss application server, to no avail....

So... does the process engine "stop looking" ???

What makes it "stop looking" at these entries in the Task Persistence DB ?

Jimmy,

  If Hornet Queue cannot process the message, it will never leave task persistence.

  The error posted here does not look like jboss 6, is it? you are using an RDB, I can see that.

  You can always try truncating the data in the message tables, then restart.

   Simply restarting may not resolve the issue.

   This part of the error should point you in the correct direction:

   tried '3' times, exception getting object id: The connection is closed.

    This tells us that jboss is having connection pooling/timeout issues.

    I would solve that issue first and get rid of those errors, as we know that those types of errors will stop tasks from processing.

    You may  be facing more that one issue at a time.

    You do have a support ticket opened. so we should not duplicate work.

Thanks,

Bill Patton

I have updated the support ticket with the information from this post.

Thanks... and I am asking the community if anyone else has experienced these issues, and what they did to resolve them.

Yes sir, I understand... That is what communities is all about.

I was just letting you know I was stepping out.

Good luck.