-
Bug
-
Resolution: Done
-
Minor
-
2.4.1
-
None
From Dimitris:
some times, when shutting down the 'all config I see something like that.
Any ideas?
14:08:03,951 INFO [TomcatDeployer] undeploy, ctxPath=/jbossmq-httpil, warUrl=
./deploy-hasingleton/jms/jbossmq-httpil.sar/jbossmq-httpil.war/
14:08:04,282 INFO [TreeCache] stopService(): closing the channel
14:08:04,352 ERROR [UDP] failed handling data from 192.168.0.22:4864
QueueClosedException
at org.jgroups.util.Queue.add(Queue.java:139)
at org.jgroups.protocols.TP.receive(TP.java:777)
at org.jgroups.protocols.UDP$UcastReceiver.run(UDP.java:961)
at java.lang.Thread.run(Thread.java:595)
14:08:04,352 INFO [TreeCache] stopService(): stopping the dispatcher
AFAICT, problem is that UDP.UcastReceiver.stop() does not wait for the receiver thread to terminate it's loop before returning. Therefore the shutdown thread can go on and close the IncomingPacketHandler, leading to the QueueClosedException if UcastReceiver is in the middle of its loop and has read another packet off the wire.
Possible solution is to have the shutdown thread join UcastReceiver.thread in UcastReceiver.stop(). Give UcastReceiver.thread time to terminate its loop before continuing on with the shutdown.
- relates to
-
JGRP-428 Backport of http://jira.jboss.com/jira/browse/JGRP-422
-
- Resolved
-