I found the reason for this behavior. I have repeat timers in my process definition. The timeout event listeners for those timers check for a particular condition and if it is met, transition to a new activity. What's happening in that case is for some reason, though the timer is getting deleted from the database, the timer id is still being used by DispatcherThread class. It passes along this id to the ExecuteJobCmd which throws the exception I posted above as it can't find any timer in the database by that id. It retries upto 3 times and then gives up. That's why I see the exception throws 3 times in the logs. So is there a better way we could turn off the repeat timers when transitioning to a new activity? View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4263589#... Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&a...