10:01 a.m.
concurrent job acquirement fails
--------------------------------
Key: JBPM-974
URL: http://jira.jboss.com/jira/browse/JBPM-974
Project: JBoss jBPM
Issue Type: Bug
Components: Core Engine
Affects Versions: jBPM jPDL 3.2
Reporter: Alexander Schlett
Assigned To: Tom Baeyens
when more than 1 job executor thread is configured and concurrent async nodes are to be
executed, it results in hibernate exceptions.
reason: job acquirement is not synchronized well, resulting in multiple threads trying to
work on the same job.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
10:26 a.m.
New subject: [JBoss JIRA] Commented: (JBPM-974) concurrent job acquirement fails
[ http://jira.jboss.com/jira/browse/JBPM-974?page=comments#action_12363386 ]
Edward Staub commented on JBPM-974:
-----------------------------------
Jira doesn't show the context of my last comment... I added the "is related
to" link to JBPM-626, and added the comment regarding that link:
"I'm not sure these are related... CommandExecutorThread seems to be refactored
out of existence, and I'm too lazy to figure out where it went to."
concurrent job acquirement fails
--------------------------------
Key: JBPM-974
URL: http://jira.jboss.com/jira/browse/JBPM-974
Project: JBoss jBPM
Issue Type: Bug
Components: Core Engine
Affects Versions: jBPM jPDL 3.2
Reporter: Alexander Schlett
Assigned To: Tom Baeyens
when more than 1 job executor thread is configured and concurrent async nodes are to be
executed, it results in hibernate exceptions.
reason: job acquirement is not synchronized well, resulting in multiple threads trying to
work on the same job.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
8:31 a.m.
New subject: [JBoss JIRA] Closed: (JBPM-974) concurrent job acquirement fails
[ http://jira.jboss.com/jira/browse/JBPM-974?page=all ]
Tom Baeyens closed JBPM-974.
----------------------------
Resolution: Cannot Reproduce Bug
concurrent job acquirement fails
--------------------------------
Key: JBPM-974
URL: http://jira.jboss.com/jira/browse/JBPM-974
Project: JBoss jBPM
Issue Type: Bug
Components: Core Engine
Affects Versions: jBPM jPDL 3.2
Reporter: Alexander Schlett
Assigned To: Tom Baeyens
when more than 1 job executor thread is configured and concurrent async nodes are to be
executed, it results in hibernate exceptions.
reason: job acquirement is not synchronized well, resulting in multiple threads trying to
work on the same job.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
8:31 a.m.
New subject: [JBoss JIRA] Commented: (JBPM-974) concurrent job acquirement fails
[ http://jira.jboss.com/jira/browse/JBPM-974?page=comments#action_12363872 ]
Tom Baeyens commented on JBPM-974:
----------------------------------
the forum thread doesn't give my much to go on. the stack trace is a 'normal'
situation in a clustered environment. the idea is that the first transaction that commits
wins and the other one is rolled back. so the job should remain in the queue and get
picked up for execution again by the same thread for a next try.
please reopen if you have more concrete problem report. the job executor is tested witih
5 concurrent threads (but they are running on the same machine), which should not be a
problem.
concurrent job acquirement fails
--------------------------------
Key: JBPM-974
URL: http://jira.jboss.com/jira/browse/JBPM-974
Project: JBoss jBPM
Issue Type: Bug
Components: Core Engine
Affects Versions: jBPM jPDL 3.2
Reporter: Alexander Schlett
Assigned To: Tom Baeyens
when more than 1 job executor thread is configured and concurrent async nodes are to be
executed, it results in hibernate exceptions.
reason: job acquirement is not synchronized well, resulting in multiple threads trying to
work on the same job.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
8:46 a.m.
New subject: [JBoss JIRA] Commented: (JBPM-974) concurrent job acquirement fails
[ http://jira.jboss.com/jira/browse/JBPM-974?page=comments#action_12363875 ]
Alexander Schlett commented on JBPM-974:
----------------------------------------
Ok, if it is normal behavior to try to acquire a job and throw an Exception if that fails,
then indeed there's no problem. Because JobExecutor threads are started
simultaneously they also simultaneously try to acquire jobs. I changed that by
synchronizing the acquirement on a static variable, so that only one JobExecutor thread
gets the job (they acquire jobs now one by one). This works well for us and no Exceptions
show up. Of course this works in a single VM only, in a cluster there still could be a
collision. The point is, an Exception should indicate an error and not pollute the logs if
it's not an error.
concurrent job acquirement fails
--------------------------------
Key: JBPM-974
URL: http://jira.jboss.com/jira/browse/JBPM-974
Project: JBoss jBPM
Issue Type: Bug
Components: Core Engine
Affects Versions: jBPM jPDL 3.2
Reporter: Alexander Schlett
Assigned To: Tom Baeyens
when more than 1 job executor thread is configured and concurrent async nodes are to be
executed, it results in hibernate exceptions.
reason: job acquirement is not synchronized well, resulting in multiple threads trying to
work on the same job.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
9:06 a.m.
New subject: [JBoss JIRA] Commented: (JBPM-974) concurrent job acquirement fails
[ http://jira.jboss.com/jira/browse/JBPM-974?page=comments#action_12363884 ]
Tom Baeyens commented on JBPM-974:
----------------------------------
there you got a point. the stack trace should not show up in the logs. an info message
would be good though so that in the logs, there is a trace of the collision and of the
fact that one of the transactions rolled back
concurrent job acquirement fails
--------------------------------
Key: JBPM-974
URL: http://jira.jboss.com/jira/browse/JBPM-974
Project: JBoss jBPM
Issue Type: Bug
Components: Core Engine
Affects Versions: jBPM jPDL 3.2
Reporter: Alexander Schlett
Assigned To: Tom Baeyens
when more than 1 job executor thread is configured and concurrent async nodes are to be
executed, it results in hibernate exceptions.
reason: job acquirement is not synchronized well, resulting in multiple threads trying to
work on the same job.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
9:10 a.m.
New subject: [JBoss JIRA] Commented: (JBPM-974) concurrent job acquirement fails
[ http://jira.jboss.com/jira/browse/JBPM-974?page=comments#action_12363885 ]
Tom Baeyens commented on JBPM-974:
----------------------------------
hmmm.. this is in the code
} finally {
jbpmContext.close();
}
acquiredJobs = jobsToLock;
log.debug("obtained locks on following jobs: "+acquiredJobs);
} catch (StaleStateException e) {
log.debug("couldn't acquire lock on job(s): "+jobsToLock);
}
and StaleStateException is the superclass of StaleObjectStateException... so i don't
yet see where this goes wrong. or maybe i already fixed it and don't remember.
concurrent job acquirement fails
--------------------------------
Key: JBPM-974
URL: http://jira.jboss.com/jira/browse/JBPM-974
Project: JBoss jBPM
Issue Type: Bug
Components: Core Engine
Affects Versions: jBPM jPDL 3.2
Reporter: Alexander Schlett
Assigned To: Tom Baeyens
when more than 1 job executor thread is configured and concurrent async nodes are to be
executed, it results in hibernate exceptions.
reason: job acquirement is not synchronized well, resulting in multiple threads trying to
work on the same job.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
6184
days inactive
6187
days old
6 comments
3 participants
participants (3)
-
Alexander Schlett (JIRA) -
Edward Staub (JIRA)
-
Tom Baeyens (JIRA)