4 Replies Latest reply on Jun 28, 2011 3:35 AM by ronsen

    SFSB State replication error

    ronsen

      Hey all,

       

      whenever i try that for example on failover a client session can be continued with another server with a sfsb, it throws me an error, that stateful session bean 123489-123807-123789-18238 couldnt be found.

       

      SFSB replication across the cluster is handled by the container actually and should be autoamtical right?

        • 1. Re: SFSB State replication error
          ronsen

          Here, thats the exact error that appears on failover:

           

           

          javax.ejb.NoSuchEJBException: Could not find stateful bean: 4sk1u3m-z86s60-gpfi8z4m-1-gpfiaoso-dl

           

          Do I have to change any settings to make that work? i already deleted all temp/tmp/work/etc. folders to make sure everythings in the correct state, just in case...

           

           

          Thats what i use in my bean:

           

          @Stateful(mappedName="ejb/FailoverFirstAvailableBean")

          @Clustered(loadBalancePolicy="FirstAvailable")

           

          @CacheConfig(name="slsb/dist")

           

           

           

          and thats the lookup from the client if its important:

           

          Properties env = 
          new Properties();env.put(Context.
          INITIAL_CONTEXT_FACTORY,"org.jnp.interfaces.NamingContextFactory");env.put(Context.
          URL_PKG_PREFIXES, "org.jboss.naming:org.jnp.interfaces");String bindAddress = System.getProperty("jboss.bind.address","jnp://"+address);
          env.put(Context.
          PROVIDER_URL, bindAddress);
           
          
          ctx = new InitialContext(env);
          Failover f = (Failover) 
          ctx.lookup(bean);
          
          
           
          
          

          • 2. Re: SFSB State replication error
            ronsen

            Ok, at first i solved the rror by adjusting my stupid mistake in cacheConfig SLSB --> SFSB, but now the cluster keeps on sending:

             

            discarded message from different cluster "DefaultPartition-SFSB/jar=

             

            Why is it using a different partition after changing to sfsb

            • 3. Re: SFSB State replication error
              ronsen

              Damn, the error is totally devastating and annoying. I dont just want to adjust log4j not to show it, i actually want to know why it is there even if theres just one node and no cluster ?!

              If I make it an slsb again, it works -> SFSB annoying errors!

               

              08:01:58,017 WARN  [org.jgroups.protocols.UDP] discarded message from different cluster "DefaultPartition-SFSB/jar=CMFailover.jar,name=FailoverFirstIdenticalBean,service=EJB3" (our cluster is "DefaultPartition-SFSB/jar=CMFailover.jar,name=FailoverRan
              domRobinBean,service=EJB3"). Sender was Server-xyz-46263

               

              does anybody know why??

               

              Thanks a lot,

              • 4. Re: SFSB State replication error
                ronsen

                For some reason the @CacheConfig parameters caused the problem.

                 

                Should be resolved.