Yes, I have seen the same exceptions when using jBPM in a cluster, all instances using the same db.
according to http://jira.jboss.com/jira/browse/JBPM-995 this is expected behaviour, as jBPM synchronizes on the database using optimistic locking.
hmmmm,.... thanks for answer. So I will try it with the expected behaviour. I hope that there will not be a lot of retries (it seems to me that there is too many retries that slows down performance in my simple scenarious).
If I have some spare time, I will try pesimisistic locking but I expect deadlock issues with this type of locking.
Recently we were dealing with problems in running JobExecutor in more than 1 thread. See http://jira.jboss.org/jira/browse/JBPM-1072 for instance. Unfortunatelly, I feel like it is not reliable at the moment.
Dev guys, correct me if I'm wrong...
What "it" are you referring to?