[ http://jira.jboss.com/jira/browse/JBSEAM-2281?page=all ] Damian Harvey updated JBSEAM-2281: ---------------------------------- Description: When using incremental hot deploy (ie. WEB-INF/dev) an Event is fired as many times as it's been since you last did a full deploy. For instance, if I start up my server and cause an event to fire (eg. @RaiseEvent("hello") on a method of mine), do an incremental hot deploy, cause the event to fire again; Then I will get the event firing twice on the second fire, three times on the third etc. If I do a full undeploy/redeploy it 'resets' the number of times it fires and all is good. This doesn't affect regular deployment - only hot deployment. Reproduce by creating an observer as such: public class ObserveHello { @Observer("hello") public void printHello() { System.out.println("Hello Event has been observed"); } } public class SayHello { @RaiseEvent("hello") public void sayHello() { System.out.println("Raising Hello event"); } } Then it's just a matter of calling the sayHello() method. Seeing the event fired once. Changing something minor. Doing an incremental deploy. Calling the sayHello() method again. You should then see the event fired twice instead of the expected once. Cheers, Damian. was: When using incremental hot deploy (ie. WEB-INF/dev) an Event is fired as many times as it's been since you last did a full deploy. For instance, if I start up my server and cause an event to fire (eg. @RaiseEvent("hello") on a method of mine), do an incremental hot deploy, cause the event to fire again; Then I will get the event firing twice on the second fire, three times on the third etc. If I do a full undeploy/redeploy it 'resets' the number of times it fires and all is good. This doesn't affect regular deployment - only hot deployment. Cheers, Damian. -- 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

[ http://jira.jboss.com/jira/browse/JBSEAM-2281?page=comments#action_12389635 ] Damian Harvey commented on JBSEAM-2281: --------------------------------------- Similar behaviour is also experienced by the Quartz timer. If a job is scheduled to run and you have done a number of hot deploys, then the job will run as many times in parallel as you have done hot deploys since the last full deploy. Also, if the Observer method signature has changed when doing hot deploys you can experience a methodNotFound exception. See the forum reference. -- 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

[ http://jira.jboss.com/jira/browse/JBSEAM-2281?page=comments#action_12392102 ] Pete Muir commented on JBSEAM-2281: ----------------------------------- It also leads to leaking of class objects, as the old Method/Class references don't get released upon re-deploy, eventually leading to a perm-gen out-of-mem. -- 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

[ http://jira.jboss.com/jira/browse/JBSEAM-2281?page=all ] Pete Muir closed JBSEAM-2281. ----------------------------- Resolution: Done Fixed. Thanks for the bug report! -- 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