Bugs item #1051174, was opened at 2004-10-21 00:55 Message generated for change (Settings changed) made by ejort You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=376685&aid=1051174&group_id=22866
Category: JBossServer Group: v4.0 >Status: Closed >Resolution: Fixed Priority: 5 Submitted By: Scott M Stark (starksm) >Assigned to: Adrian Brock (ejort) Summary: Deadlock on shutdown with java5 testcase Initial Comment: There is a jboss-minimal-tests target in the testsuite ant16x-build.xml that starts the minimal config with the testsuite/output/lib/shutdown.sar deployed. This stops the server by undeploying the shutdown.sar. The ExitOnShutdown in this sar calls System.exit in its stop method and under java5 this is causing a deadlock because the javax.management.NotificationBroadcasterSupport.sendNotification method happens to be synchronized in the java5 implementation while it is not in the jboss implementation. See the attached thread dump. ---------------------------------------------------------------------- Comment By: Adrian Brock (ejort) Date: 2004-10-21 09:48 Message: Logged In: YES user_id=9459 It had already been reported as a bug and fixed for the next java 5 release: http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=5093922 ---------------------------------------------------------------------- Comment By: Adrian Brock (ejort) Date: 2004-10-21 01:46 Message: Logged In: YES user_id=9459 Another solution is to make ServiceMBeanSupport extend JBossNotificationBroadcasterSupport to avoid the synchronization. This would be a copy of the current JBossMX NotificationBroadcasterSupport and make javax.management.NotificationBroadcasterSupport delegate to it. I will try this. ---------------------------------------------------------------------- Comment By: Adrian Brock (ejort) Date: 2004-10-21 01:21 Message: Logged In: YES user_id=9459 The fix would appear to be not to synchronize on "this" or have NO synchronized methods on anything that extends NotificationBroadcasterSupport. I see nothing in the javadoc that suggests this leak of a synchronization to subclasses. It should be an internal synchronization. I will forward this to the JMX 2.0 spec commitee. But this probably needs reporting as a bug to Sun. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=376685&aid=1051174&group_id=22866 ------------------------------------------------------- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl _______________________________________________ JBoss-Development mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-development
