Connection.start sometimes fails to connect when master container is destroyed when using replicated levelDB and fabric discovery.

[mIteration=77
�[0;32m10-12-2014 17:19:35,189 | INFO | [FabricDiscoveryAgent] | call - Using local ZKClient
�[m�[0;32m10-12-2014 17:19:35,190 | INFO | [CuratorFrameworkImpl] | start - Starting
�[m�[0;32m10-12-2014 17:19:35,194 | INFO | [ZooKeeper] | <init> - Initiating client connection, connectString=localhost:2181 sessionTimeout=60000 watcher=org.apache.curator.ConnectionState@41e65439
�[m�[0;32m10-12-2014 17:19:35,195 | INFO | [ClientCnxn] | logStartConnect - Opening socket connection to server dhcp-4-224.brq.redhat.com/127.0.0.1:2181
�[m�[0;32m10-12-2014 17:19:35,196 | INFO | [ClientCnxn] | primeConnection - Socket connection established to dhcp-4-224.brq.redhat.com/127.0.0.1:2181, initiating session
�[m�[0;32m10-12-2014 17:19:35,206 | INFO | [ClientCnxn] | onConnected - Session establishment complete on server dhcp-4-224.brq.redhat.com/127.0.0.1:2181, sessionid = 0x14a34f50ff001a1, negotiated timeout = 40000
�[m�[0;32m10-12-2014 17:19:35,207 | INFO | [ConnectionStateManager] | postState - State change: CONNECTED
�[m�[0;32m10-12-2014 17:19:35,794 | INFO | [SSHClient] | executeCommand - Command: activemq:list
�[m�[0;32m10-12-2014 17:19:35,997 | INFO | [SSHClient] | executeCommand - Response: brokerName = broker-nossl
 
 
�[m!!!Restart finished container 0
!!!Stopping container 1
!!!!Killing container broker-nossl-c1
�[0;32m10-12-2014 17:19:35,997 | INFO | [SSHClient] | executeCommand - Command: shell:exec kill -9 21056
�[m�[0;32m10-12-2014 17:19:36,237 | INFO | [DiscoveryTransport] | onServiceAdd - Adding new broker connection URL: tcp://dhcp-10-40-3-26.brq.redhat.com:39445
�[m�[0;32m10-12-2014 17:19:36,241 | INFO | [FailoverTransport] | doReconnect - Successfully connected to tcp://dhcp-10-40-3-26.brq.redhat.com:39445
�[m�[0;33m10-12-2014 17:19:36,286 | WARN | [FailoverTransport] | handleTransportFailure - Transport (tcp://dhcp-10-40-3-26.brq.redhat.com/10.40.3.26:39445@54170) failed, attempting to automatically reconnect
org.apache.activemq.transport.InactivityIOException: Cannot send, channel has already failed: tcp://10.40.3.26:39445
        at org.apache.activemq.transport.AbstractInactivityMonitor.doOnewaySend(AbstractInactivityMonitor.java:297)
        at org.apache.activemq.transport.AbstractInactivityMonitor.oneway(AbstractInactivityMonitor.java:286)
        at org.apache.activemq.transport.TransportFilter.oneway(TransportFilter.java:85)
        at org.apache.activemq.transport.WireFormatNegotiator.oneway(WireFormatNegotiator.java:104)
        at org.apache.activemq.transport.failover.FailoverTransport.oneway(FailoverTransport.java:665)
        at org.apache.activemq.transport.TransportFilter.oneway(TransportFilter.java:85)
        at org.apache.activemq.transport.MutexTransport.oneway(MutexTransport.java:68)
        at org.apache.activemq.transport.ResponseCorrelator.asyncRequest(ResponseCorrelator.java:81)
        at org.apache.activemq.transport.ResponseCorrelator.request(ResponseCorrelator.java:86)
        at org.apache.activemq.ActiveMQConnection.syncSendPacket(ActiveMQConnection.java:1406)
        at org.apache.activemq.ActiveMQConnection.ensureConnectionInfoSent(ActiveMQConnection.java:1522)
        at org.apache.activemq.ActiveMQConnection.start(ActiveMQConnection.java:527)
        at org.jboss.fuse.qa.amq.threads.ConnectionRestarter.performAction(ConnectionRestarter.java:19)
        at org.jboss.fuse.qa.amq.threads.AbstractSenderReceiver.run(AbstractSenderReceiver.java:63)
        at java.lang.Thread.run(Thread.java:744)
�[m�[0;32m10-12-2014 17:19:36,470 | INFO | [SSHClient] | executeCommand - Response:
�[m�[0;32m10-12-2014 17:19:36,471 | INFO | [Container] | killContainerAtLocalHost - Result of killing master process :
�[m�[0;32m10-12-2014 17:19:36,471 | INFO | [SSHClient] | executeCommand - Command: shell:exec kill -9 21056
�[m�[0;32m10-12-2014 17:19:36,780 | INFO | [SSHClient] | executeCommand - Response:
�[m�[0;32m10-12-2014 17:19:36,781 | INFO | [Container] | killContainerAtLocalHost - Result of killing master process :
�[m!!!!Waiting for empty process broker-nossl-c1

mq-create --no-ssl --parent-profile=mq-replicated --group a broker-nossl
container-create-child --profile mq-broker-a.broker-nossl root broker-nossl-c1
container-create-child --profile mq-broker-a.broker-nossl root broker-nossl-c2
container-create-child --profile mq-broker-a.broker-nossl root broker-nossl-c3

discovery:(fabric://a)?useExponentialBackoff=false&trace=true

Connection con = getConnectionFactory().createConnection();
		con.start();
		Session session = con.createSession(false, Session.AUTO_ACKNOWLEDGE);
		Thread.sleep(10);
		session.close();
		con.stop();
		con.close();