[timfox]

Actually Jeff, I think you were right: The SESSION_CLOSE won't suffice as a LAST message since many sessions can share the same connection, so I think you need a CLOSE message on the connection.

[clebert.suconic]

This is still happening....

To replicate:


ant runServer



ant runListener

Ctrl-C

ant runListener


ant runSender



The listener hangs:

 [java] 07:32:42,926 INFO [org.jboss.core.example.PerfSender] received 96000 messages in 9.66s
 [java] 07:32:43,080 INFO [org.jboss.core.example.PerfSender] received 98000 messages in 9.81s

It doesn't receive all the messages.

[clebert.suconic]

I was running this on my branch created two days ago.


Unless someone fixed it past 2 days. I will validate it on trunk later today.

[clebert.suconic]

Yep.. it also happens on trunk as I expected.

[jmesnil]

Yep.. it also happens on trunk as I expected.

did you wait for the connection TTL to be hit (5 minutes)?

[jmesnil]

I tried it and the connection TTL is effectively hit but the messages sent to the 1st listener's buffer are not put back in the queue.

 [java] 14:13:38,811 WARNING [org.jboss.messaging.core.remoting.impl.RemotingConnectionImpl] Connection failure has been detected Did not receive ping on connection. It is likely a client has exited or crashed without closing its connection, or the network between the server and client has failed. The connection will now be closed.:3
 [java] 14:13:38,811 WARNING [org.jboss.messaging.core.server.impl.ServerSessionImpl] Client connection failed, clearing up resources for session 4fbd267a-4ea5-11de-b784-b725ab0e15cf
 [java] java.lang.Exception
 [java] at org.jboss.messaging.core.server.impl.ServerSessionImpl.connectionFailed(ServerSessionImpl.java:1157)
 [java] at org.jboss.messaging.core.remoting.impl.RemotingConnectionImpl.callFailureListeners(RemotingConnectionImpl.java:582)
 [java] at org.jboss.messaging.core.remoting.impl.RemotingConnectionImpl.fail(RemotingConnectionImpl.java:461)
 [java] at org.jboss.messaging.core.remoting.server.impl.RemotingServiceImpl$FailedConnectionsTask.run(RemotingServiceImpl.java:374)
 [java] at java.util.TimerThread.mainLoop(Timer.java:512)
 [java] at java.util.TimerThread.run(Timer.java:462)
 [java] 14:13:38,813 WARNING [org.jboss.messaging.core.server.impl.ServerSessionImpl] Cleared up resources for session 4fbd267a-4ea5-11de-b784-b725ab0e15cf

[jmesnil]

Yep.. it also happens on trunk as I expected.

The code is good. It happens in the trunk because the server *pre-acks* the messages it delivers to the perfListener.

If I set pre-ack to false, the messages are all received by the 2nd consumer when the connection TTL is hit and the server clears up the 1st consumer resources.