]
Brian Stansberry updated JBCLUSTER-234:
---------------------------------------
Component/s: (was: HA-Client)
(was: HA-Server-API)
(was: HA-Server-Cache-SPI)
(was: HA-Server-Cache-JBC)
(was: HA-Server-Impl)
Priority: Major (was: Critical)
Priority field is for use by committers only.
JBoss MQ Topic and Queues deployed on both the servers in cluster.
------------------------------------------------------------------
Key: JBCLUSTER-234
URL:
https://jira.jboss.org/jira/browse/JBCLUSTER-234
Project: JBoss Clustering
Issue Type: Bug
Security Level: Public(Everyone can see)
Environment: Server: JBoss AS 4.2.2.
Components Used: HA Singleton, HA JBoss MQ, EJB 3.0
JVM: 96GB Virtual Memory, Concurrent Garbage Collection algorithm.
Cluster: 2 servers connected with TCPPING mechanism.
Reporter: Gideon Raj
Assignee: Brian Stansberry
When both the servers are up and running, one pause (either cause by a GC or a network
glitch) causes server2 to start the HA Singleton object and also bind the JMS Queues and
Topics. At this point both the servers are running as primaries.
(Note regarding the pause: We are monitoring the GC collections using JConsole. Only the
ParNew collections have taken place, the ConcurrentMarkSweep collection has not yet taken
place).
After a couple of minutes, both the servers communicate, and server1 reestablishes as the
sole primary in the cluster. But the HA Singleton, JMS Queues and Topics created in
server2 remains and is not unbound.
The sequence of events:
Log message from server1:
2009-04-22 17:16:11,251
org.jgroups.blocks.BasicConnectionTable$Connection._send(BasicConnectionTable.java:443)
WARN [org.jgroups.blocks.ConnectionTable] peer closed connection, trying to re-send msg
2009-04-22 17:16:11,252
org.jgroups.blocks.BasicConnectionTable$Connection._send(BasicConnectionTable.java:449)
ERROR [org.jgroups.blocks.ConnectionTable] 2nd attempt to send data failed too
Log message from server2
2009-04-22 17:15:42,828
org.jboss.ha.framework.server.HAPartitionImpl.suspect(HAPartitionImpl.java:584) INFO
[org.jboss.ha.framework.interfaces.HAPartition.CUSTOMPartition] Suspected member:
server1:7800
....
2009-04-22 17:15:42,940
org.jboss.ha.framework.server.DistributedReplicantManagerImpl.membershipChanged(DistributedReplicantManagerImpl.java:374)
INFO [org.jboss.ha.framework.server.DistributedReplicantManagerImpl.CUSTOMPartition] Dead
members: 1 ([server1:1199])
2009-04-22 17:15:42,941
org.jboss.ha.framework.server.DistributedReplicantManagerImpl.membershipChanged(DistributedReplicantManagerImpl.java:375)
INFO [org.jboss.ha.framework.server.DistributedReplicantManagerImpl.CUSTOMPartition] New
Members : 0 ([])
2009-04-22 17:15:42,941
org.jboss.ha.framework.server.DistributedReplicantManagerImpl.membershipChanged(DistributedReplicantManagerImpl.java:376)
INFO [org.jboss.ha.framework.server.DistributedReplicantManagerImpl.CUSTOMPartition] All
Members : 1 ([server2:1199])
.....
2009-04-22 17:16:11,255
org.jgroups.protocols.pbcast.NAKACK.handleMessage(NAKACK.java:668) WARN
[org.jgroups.protocols.pbcast.NAKACK] server2:49665] discarded message from non-member
server1:57210, my view is [server2:49665|2] [server2:49665]
2009-04-22 17:16:11,259
org.jgroups.protocols.pbcast.NAKACK.handleMessage(NAKACK.java:668) WARN
[org.jgroups.protocols.pbcast.NAKACK] server2:49663] discarded message from non-member
server1:57208, my view is [server2:49663|2] [server2:49663]
.....
2009-04-22 17:16:13,877 org.jboss.cache.TreeCache.viewAccepted(TreeCache.java:5687) INFO
[org.jboss.cache.TreeCache] viewAccepted(): [server2:49663|3] [server2:49663,
server1:57824]
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: