HTTP Session Tomcat / JBoss Issue

[ben.pienig]

Hello All,

I'm having problems with my setup of clustering.

Here is the information, and the problem

2 JBOSS Nodes Running 4.0.5.GA

10:19:37,494 INFO [Server] Starting JBoss (MX MicroKernel)...
10:19:37,496 INFO [Server] Release ID: JBoss [Zion] 4.0.5.GA (build: CVSTag=Branch_4_0 date=200610162339)

An apache frontend running.. 2.2.4

Server version: Apache/2.2.4 (Unix)
Server built: Jul 1 2007 18:11:30

Apache frontend is running Mod JK 1.2.23

Module Name: mod_jk-1.2.23-apache-2.2.x.so

httpd.conf
-----------

<IfModule mod_jk.c>
 JkLogFile /var/log/httpd/modjk.log
 JkLogLevel info
 JkWorkersFile /etc/httpd/jkworkers.properties
 JkShmFile /var/log/httpd/jk.shm
 JkOptions +ForwardKeySize +ForwardURICompatUnparsed -ForwardDirectories
 LogFormat "%h %l %u %t \"%r\" %>s %b %{JK_WORKER_NAME}n %{JK_LB_FIRST_NAME}n %{JK_LB_FIRST_BUSY}n %{JK_LB_LAST_NAME}n %{JK_LB_LAST_BUSY}n" mod_jk_log
 CustomLog /var/log/httpd/modjk_access.log mod_jk_log

 # JK Balancer manager
 <VirtualHost *:80>
 ServerAdmin [e-mail]
 DocumentRoot /srv/httpd/htdocs
 JkMount /* jboss
 </VirtualHost>

 <VirtualHost *:5555>
 ServerAdmin [e-mail]
 DocumentRoot /srv/httpd/htdocs
 <Location /status/>
 JkMount jkstatus
 </Location>
 </VirtualHost>
</IfModule>

jkworkers.properties
-----------------------

worker.list=jboss,jkstatus

# First Node
worker.node1.type=ajp13
worker.node1.host=192.168.7.235
worker.node1.port=7929
worker.node1.lbfactor=1
worker.node1.socket_timeout=10
worker.node1.connect_timeout=20000

# Second Node
worker.node2.type=ajp13
worker.node2.host=192.168.7.236
worker.node2.port=7929
worker.node2.lbfactor=1
worker.node2.socket_timeout=10
worker.node2.connect_timeout=20000

# Load balancer
worker.jboss.type=lb
worker.jboss.balance_workers=node1,node2
worker.jboss.sticky_session=1

worker.jkstatus.type=status

tc5-cluster.sar/META-INF/jboss-service.xml

<attribute name="ClusterName">Tomcat-${jboss.partition.name:Cluster}</attribute>
<attribute name="IsolationLevel">REPEATABLE_READ</attribute>
<attribute name="CacheMode">REPL_ASYNC</attribute>

-----------------------

Here's the issue.

I start up the cluster, and all works fine. The moment I start putting any load onto it I get the following problem....

10:25:41,060 INFO [CacheListener] Possible concurrency problem: Replicated version id 22 matches in-memory version for session ml5CzYIuE0X62k2vNReP4g**
10:25:59,215 INFO [CacheListener] Possible concurrency problem: Replicated version id 10 matches in-memory version for session GXDm0EZryMgDh-e17hE2Eg**
10:26:04,487 INFO [CacheListener] Possible concurrency problem: Replicated version id 11 matches in-memory version for session GXDm0EZryMgDh-e17hE2Eg**
10:26:04,508 WARN [MsgContext] Error sending end packet
java.net.SocketException: Broken pipe
 at java.net.SocketOutputStream.socketWrite0(Native Method)
 at java.net.SocketOutputStream.socketWrite(SocketOutputStream.java:92)
 at java.net.SocketOutputStream.write(SocketOutputStream.java:136)
 at org.apache.jk.common.ChannelSocket.send(ChannelSocket.java:531)
 at org.apache.jk.common.JkInputStream.endMessage(JkInputStream.java:112)
 at org.apache.jk.core.MsgContext.action(MsgContext.java:293)
 at org.apache.coyote.Response.action(Response.java:182)
 at org.apache.coyote.Response.finish(Response.java:304)
 at org.apache.jk.server.JkCoyoteHandler.invoke(JkCoyoteHandler.java:204)
 at org.apache.jk.common.HandlerRequest.invoke(HandlerRequest.java:282)
 at org.apache.jk.common.ChannelSocket.invoke(ChannelSocket.java:767)
 at org.apache.jk.common.ChannelSocket.processConnection(ChannelSocket.java:697)
 at org.apache.jk.common.ChannelSocket$SocketConnection.runIt(ChannelSocket.java:889)
 at org.apache.tomcat.util.threads.ThreadPool$ControlRunnable.run(ThreadPool.java:684)
 at java.lang.Thread.run(Thread.java:595)
10:26:04,512 WARN [ChannelSocket] processCallbacks status 2
10:31:06,604 INFO [CacheListener] Possible concurrency problem: Replicated version id 4 matches in-memory version for session ml5CzYIuE0X62k2vNReP4g**
10:31:10,643 INFO [CacheListener] Possible concurrency problem: Replicated version id 5 matches in-memory version for session ml5CzYIuE0X62k2vNReP4g**

Does anyone have any idea why this problem is occuring?

Any solutions to this problem, is an issue with my configuration?

Thanks for the help in Advance,

Ben

[ben.pienig]

It looks like I solved the problem.....

Both of the JBOSS Clusters jvmRoutes were set to the same routeId.

Once this was changed, the systems seemed to function fine.

10:31:10,643 INFO [CacheListener] Possible concurrency problem: Replicated version id 5 matches in-
memory version for session ml5CzYIuE0X62k2vNReP4g**

If you see this, check your JBOSS_HOME/server/all/deploy/jbossweb-tomcat50.sar/server.xml jvmRoute settings.

Ben

[brian.stansberry]

Thanks for posting the answer. :-)

If you see that "INFO [CacheListener] Possible concurrency problem: Replicated version id XX matches in-memory version for session YYYYYYYYYYYY**" logging it's normally a sign of sticky sessions not working properly. Having the wrong jvmRoute would cause a breakdown in sticky sessions.