jboss 4 redeploy problem

[Huiying Duan]

Hi,

I have met a problem when redeploying the ear artifact on my jboss4 cluster. The jboss cluster has 3 nodes. In the artifact, there is a mbean which uses a distributed list created by hazelcast.

Here is the error information:

2014-03-17 10:16:35,995 INFO  [com.hazelcast.partition.PartitionStateOperation] [172.19.60.2]:5701 [dev] Can't overwrite cause
java.lang.IllegalStateException: Can't overwrite cause
        at java.lang.Throwable.initCause(Throwable.java:320)
        at org.jboss.mx.loading.RepositoryClassLoader.findClass(RepositoryClassLoader.java:624)
        at java.lang.ClassLoader.loadClass(ClassLoader.java:306)
        at org.jboss.mx.loading.RepositoryClassLoader.loadClassImpl(RepositoryClassLoader.java:474)
        at org.jboss.mx.loading.RepositoryClassLoader.loadClass(RepositoryClassLoader.java:415)
        at java.lang.ClassLoader.loadClass(ClassLoader.java:247)
        at com.hazelcast.partition.PartitionServiceImpl.clearPartitionReplica(PartitionServiceImpl.java:595)
        at com.hazelcast.partition.PartitionServiceImpl.access$500(PartitionServiceImpl.java:48)
        at com.hazelcast.partition.PartitionServiceImpl$LocalPartitionListener.replicaChanged(PartitionServiceImpl.java:134)
        at com.hazelcast.partition.PartitionImpl.setReplicaAddress(PartitionImpl.java:69)
        at com.hazelcast.partition.PartitionImpl.setPartitionInfo(PartitionImpl.java:88)
        at com.hazelcast.partition.PartitionServiceImpl.processPartitionRuntimeState(PartitionServiceImpl.java:457)
        at com.hazelcast.partition.PartitionStateOperation.run(PartitionStateOperation.java:53)
        at com.hazelcast.spi.impl.OperationServiceImpl.doRunOperation(OperationServiceImpl.java:273)
        at com.hazelcast.spi.impl.OperationServiceImpl.access$800(OperationServiceImpl.java:51)
        at com.hazelcast.spi.impl.OperationServiceImpl$RemoteOperationProcessor.run(OperationServiceImpl.java:736)
        at com.hazelcast.util.executor.ManagedExecutorService$Worker.run(ManagedExecutorService.java:166)
        at java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecutor.java:886)
        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:908)
        at java.lang.Thread.run(Thread.java:662)
        at com.hazelcast.util.executor.PoolExecutorThreadFactory$ManagedThread.run(PoolExecutorThreadFactory.java:59)


Members [6] {
        Member [172.19.60.52]:5701
        Member [172.19.60.3]:5701
        Member [172.19.60.2]:5701
        Member [172.19.60.2]:5702 this
        Member [172.19.60.3]:5702
        Member [172.19.60.52]:5702
}

Another question: how can I get rid of the useless members like Member [172.19.60.3]:5701? I think there are three mbean iinstances which are obsolete.

Best,
Huiying

[Peter Veentjer]

Which Hazelcast version are you using.

[Huiying Duan]

3.2-RC2

[Peter Veentjer]

The stacktrace doesn't correspond to code in Hazelcast 3.2-RC2 but to 3.2-RC1

It looks like the exception is not caused within the Hazelcast code, but in

    at org.jboss.mx.loading.RepositoryClassLoader.findClass(RepositoryClassLoader.java:624) 

So I guess this is not related to Hazelcast, but to JBoss.

And you should shut down members if you don't want to have them. Apparently they are still running.

--
You received this message because you are subscribed to the Google Groups "Hazelcast" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hazelcast+...@googlegroups.com.
To post to this group, send email to haze...@googlegroups.com.
Visit this group at http://groups.google.com/group/hazelcast.

To view this discussion on the web visit https://groups.google.com/d/msgid/hazelcast/67b77cdd-a36a-414b-8e49-0706acc70952%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[Huiying Duan]

You are right! The problem is at jboss side. We are using a customized jboss 4 version, where it contains another hazelcast version (1.9.2.3). That is why it shows the exception when redeploying the artifact.

Regarding shutting down members, I would like to shut it down when the corresponding artifact is undeployed. Because I think it is natural that when I redeploy the artifact, I don't want the number of members doubles but stays. How can I do it?

[Peter Veentjer]

Perhaps you could do it using a ServletContextListener and kill the HazelcastInstance that is created by your web application.

PS: Hazelcast 1.x and 3.x can't be mixed. Please keep that in mind.

To view this discussion on the web visit https://groups.google.com/d/msgid/hazelcast/fe210b03-eede-442f-b7bf-d76e1a6483e4%40googlegroups.com.

[Huiying Duan]

Thanks a lot for your help, :-)