8 Replies Latest reply: Jul 23, 2012 11:01 AM by 908220 RSS

    Failover not working in a two managed server cluster environment

    908220
      Hello everyone,

      I have set up a cluster with two managed server, server A and server B. I have configured node manager, in-memory replication, replication groups, proxy server to forward requests to the cluster, and deployed an aplication to the cluster. The application has serializable objects to achieve replication process.
      When I have the two servers running it seems that the replication works fine, as the tab Cluster->Monitoring->Overview shows that i have a session in server A, and its replica in the server B. But when i kill Server A, server B can't take the session replica and it shows Http Error 500. However, when i have server A RUNNING, server B SHUTDOWN, i create a session in the aplication and after that I START the server B. When I check the replica is in the server B, I kill server A, and the failover occurs perfect. I don't know what to do in order to make failover work when the two servers of the cluster are running.

      I would appreciate your help.

      Thanks
        • 1. Re: Failover not working in a two managed server cluster environment
          Kalyan Pasupuleti-Oracle
          Hi Erick,

          Please check your logs might session or object are not seralized.

          look for the logs to verify the issue.

          Regards,
          Kal
          • 2. Re: Failover not working in a two managed server cluster environment
            908220
            Hi Kaylan,

            I've already checked instances, access, and domain logs, but the exception BEA-000126, for objects that cannot be serialized, is not being written so I think that serilization is OK. Is there a way to check the values of a replicate session in order to compare if all the objects are being serialized?. I also checked the cookies generated in the client side and it detects the primary and secondary server in the form: jsessionid!primary_server!secondary_server
            • 3. Re: Failover not working in a two managed server cluster environment
              Kalyan Pasupuleti-Oracle
              Hi Erick,

              Yes there is possibly such case with JsessionID.

              some time JsessionID is get conflict or lost.

              Can you paste me the log stack to verify the issue.

              Regards,
              kal
              • 4. Re: Failover not working in a two managed server cluster environment
                908220
                Hello Kalyan,

                Thank you for your response, this is the out file in one of the managed servers:

                ####<Jul 13, 2012 5:32:26 PM BOT> <Error> <HTTP> <serverA.localdomain> <AppInternas01> <[ACTIVE] ExecuteThread: '1' for queue: 'weblogic.kernel.Default (self-tuning)'> <<WLS Kernel>> <> <155524fda8d91764:-23c4ef5f:138823dbf32:-8000-00000000000000b2> <1342215146610> <BEA-101017> <[ServletContext@486377372[app:testapp module:/testapp path:null spec-version:3.0 version:v1.0]] Root cause of ServletException.
                javax.faces.application.ViewExpiredException: viewId:/days_off.xhtml - View /days_off.xhtml could not be restored.
                     at com.sun.faces.lifecycle.RestoreViewPhase.execute(RestoreViewPhase.java:205)
                     at com.sun.faces.lifecycle.Phase.doPhase(Phase.java:101)
                     at com.sun.faces.lifecycle.RestoreViewPhase.doPhase(RestoreViewPhase.java:116)
                     at com.sun.faces.lifecycle.LifecycleImpl.execute(LifecycleImpl.java:118)
                     at javax.faces.webapp.FacesServlet.service(FacesServlet.java:593)
                     at weblogic.servlet.internal.StubSecurityHelper$ServletServiceAction.run(StubSecurityHelper.java:242)
                     at weblogic.servlet.internal.StubSecurityHelper$ServletServiceAction.run(StubSecurityHelper.java:216)
                     at weblogic.servlet.internal.StubSecurityHelper.invokeServlet(StubSecurityHelper.java:132)
                     at weblogic.servlet.internal.ServletStubImpl.execute(ServletStubImpl.java:352)
                     at weblogic.servlet.internal.TailFilter.doFilter(TailFilter.java:25)
                     at weblogic.servlet.internal.FilterChainImpl.doFilter(FilterChainImpl.java:75)
                     at com.myenterprise.testapp.servlet.LoginFilter.doFilter(LoginFilter.java:34)
                     at weblogic.servlet.internal.FilterChainImpl.doFilter(FilterChainImpl.java:75)
                     at weblogic.servlet.internal.RequestEventsFilter.doFilter(RequestEventsFilter.java:27)
                     at weblogic.servlet.internal.FilterChainImpl.doFilter(FilterChainImpl.java:75)
                     at weblogic.servlet.internal.WebAppServletContext$ServletInvocationAction.wrapRun(WebAppServletContext.java:3288)
                     at weblogic.servlet.internal.WebAppServletContext$ServletInvocationAction.run(WebAppServletContext.java:3254)
                     at weblogic.security.acl.internal.AuthenticatedSubject.doAs(AuthenticatedSubject.java:321)
                     at weblogic.security.service.SecurityManager.runAs(SecurityManager.java:120)
                     at weblogic.servlet.provider.WlsSubjectHandle.run(WlsSubjectHandle.java:57)
                     at weblogic.servlet.internal.WebAppServletContext.doSecuredExecute(WebAppServletContext.java:2163)
                     at weblogic.servlet.internal.WebAppServletContext.securedExecute(WebAppServletContext.java:2091)
                     at weblogic.servlet.internal.WebAppServletContext.execute(WebAppServletContext.java:2074)
                     at weblogic.servlet.internal.ServletRequestImpl.run(ServletRequestImpl.java:1512)
                     at weblogic.servlet.provider.ContainerSupportProviderImpl$WlsRequestExecutor.run(ContainerSupportProviderImpl.java:255)
                     at weblogic.work.ExecuteThread.execute(ExecuteThread.java:256)
                     at weblogic.work.ExecuteThread.run(ExecuteThread.java:221)
                • 5. Re: Failover not working in a two managed server cluster environment
                  Kalyan Pasupuleti-Oracle
                  Hi Erick,

                  It look like a Bug 10155450

                  1. Please login to MOS
                  2. Click on tab "Patch & Updates"
                  3. Enter 10155450 into field "Patch Name or Number" and Search
                  4. Download the patch
                  5. Install the patch as described in README.txt


                  Regards,
                  Kal
                  • 6. Re: Failover not working in a two managed server cluster environment
                    908220
                    Hi Kalyan,

                    Thank you for yor reply, but I forgot to mention that the version of WebLogic I am using is 12c, and the patch for bug 10155450 is for 10.3.4 release.
                    I'm not sure if there is a similar bug for 12c version and a patch to be appliad.

                    Thanks in advanced.

                    Regards,

                    Erick
                    • 7. Re: Failover not working in a two managed server cluster environment
                      Kalyan Pasupuleti-Oracle
                      Hi Erick,

                      This BUG 10155450 is fixed in WLS 12.1.

                      So I think you can raise ticket with weblogic support to get patch for 12cGA mean 12.0


                      Hope this will help you.

                      Regards,
                      Kal
                      • 8. Re: Failover not working in a two managed server cluster environment
                        908220
                        Hi Kalyan,

                        Thanks for your response, it helped me a lot and I'll raise a ticket.

                        Regards,

                        Erick