geronimo-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Trygve Hardersen <try...@jotta.no>
Subject Re: 2.2 in production
Date Mon, 02 Nov 2009 14:52:40 GMT
Hi Gianny

Many thanks for you help!

I'll try to make a Geronimo build that uses this WADI version the next time
we update our servers. Will let you know what I find.

The session size in our application should be small as only a couple of
simple objects are put there, though I've not really verified this. I would
not say the servers are not under any significant load when the messages
start appearing, but again this is hard to quantify. CPU and bandwidth is
not constrained at least.

Trygve

On Sun, Nov 1, 2009 at 10:53 AM, Gianny Damour <
gianny.damour@optusnet.com.au> wrote:

> Hi Trygve,
>
> Thanks for reporting these problems with comprehensive log messages; they
> were very helpful to diagnose.
>
> The bug causing the inability to reliably restart without downtime has been
> identified and fixed. I just deployed new WADI 2.2-SNAPSHOT artifacts, which
> can be found there:
>
> http://snapshots.repository.codehaus.org/org/codehaus/wadi/
>
>
> Can you please confirm that this is now working as expected within your
> environment?
>
>
> Regarding the warning message
>
>
>  16:22:43,524 WARN  [UpdateReplicationCommand] Update has not been properly
>> cascaded due to a communication failure. If a targeted node has been lost,
>> state will be re-balanced automatically.
>>
>
> the timeout is not exposed via API. Even if this warning message is
> displayed, an update message has actually been queued for delivery to the
> relevant back-up nodes. However, there is no guarantee that in case of
> failure, the cluster will be able to restore the latest session state as the
> latest update message may not have actually been delivered to back-up nodes.
>
> I conducted load-testing on a single computer. As the network traffic going
> through the loopback interface does not actually touch the network card,
> this is a problem I was never able to observe.
>
> Do you have an idea of the traffic volume when this warning message starts
> to appear? Also, do you have an idea of the session size?
>
> Thanks,
> Gianny
>
>
> On 31/10/2009, at 3:39 AM, Trygve Hardersen wrote:
>
>  Hello
>>
>> We have been using Geronimo 2.2-SNAPSHOT in production for a good month
>> now, and I thought I'd share some experiences with the community, and maybe
>> get some help. We are an online backup service, check out jottabackup.comif you're
interested.
>>
>> Generally our experience has been very positive. We're using the GBean
>> framework for custom server components, the DB connection pools against
>> MySQL databases, stateless service EJBs and various MDB, and of course the
>> web tier (Jetty). Everything is running smoothly and we're very much looking
>> forward to 2.2 being released so we can "release" our own software.
>>
>> The issues we're having are related to WADI clustering with Jetty. First
>> we can't use Jetty7 because of GERONIMO-4846, so we're using Jetty6 which
>> works fine. The more serious issue is that we often can not update our
>> servers without downtime. This is what happens:
>>
>> We have 2 application servers (AS-000 and AS-001) running dynamic WADI
>> HTTP session replication between them. When updating we first stop one,
>> AS-000 in this case. That works fine and the active sessions are migrated to
>> AS-001:
>>
>> 23:43:18,160 INFO  [SimpleStateManager]
>> =============================
>> New Partition Balancing
>> Partition Balancing
>>    Size [24]
>>    Partition[0] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[1] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[2] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[3] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[4] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[5] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[6] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[7] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[8] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[9] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[10] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[11] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[12] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[13] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[14] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[15] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[16] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[17] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[18] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[19] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[20] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[21] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[22] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>>    Partition[23] owned by [TribesPeer [AS-001; tcp://10.0.10.101:4000]];
>> version [3]; mergeVersion [0]
>> =============================
>>
>> 23:43:28,539 INFO  [TcpFailureDetector] Verification complete. Member
>> disappeared[org.apache.catalina.tribes.membership.MemberImpl[tcp://{10, 0,
>> 10, 100}:4000,{10, 0, 10, 100},4000, alive=41104531,id={-4 -32 54 90 -109
>> -17 65 64 -117 40 -110 -14 36 93 -12 -118 }, payload={-84 -19 0 5 115 114 0
>> 50 111 ...(423)}, command={66 65 66 89 45 65 76 69 88 ...(9)}, domain={74 79
>> 84 84 65 95 87 65 68 ...(10)}, ]]
>> 23:43:28,540 INFO  [ChannelInterceptorBase] memberDisappeared:tcp://{10,
>> 0, 10, 100}:4000
>>
>> We then update AS-000 and try to start it, but it fails to rejoin the WADI
>> cluster:
>>
>> 23:46:30,784 INFO  [ReceiverBase] Receiver Server Socket bound to:/
>> 10.0.10.100:4000
>> 23:46:30,864 INFO  [ChannelInterceptorBase] memberStart
>> local:org.apache.catalina.tribes.membership.MemberImpl[tcp://
>> 10.0.10.100:4000,10.0.10.100,4000, alive=0,id={-103 34 80 -19 68 -51 70
>> -91 -108 39 -84 65 50 50 103 -107 }, payload={-84 -19 0 5 115 114 0 50 111
>> ...(423)}, command={}, domain={74 79 84 84 65 95 87 65 68 ...(10)}, ]
>> notify:false peer:AS-000
>> 23:46:30,868 INFO  [McastService] Setting cluster mcast soTimeout to 500
>> 23:46:30,908 INFO  [McastService] Sleeping for 1000 milliseconds to
>> establish cluster membership, start level:4
>> 23:46:31,139 INFO  [ChannelInterceptorBase] memberAdded:tcp://{10, 0, 10,
>> 101}:4000
>> 23:46:31,908 INFO  [McastService] Done sleeping, membership established,
>> start level:4
>> 23:46:31,912 INFO  [McastService] Sleeping for 1000 milliseconds to
>> establish cluster membership, start level:8
>> 23:46:31,927 INFO  [BufferPool] Created a buffer pool with max
>> size:104857600 bytes of type:org.apache.catalina.tribes.io.BufferPool15Impl
>> 23:46:32,912 INFO  [McastService] Done sleeping, membership established,
>> start level:8
>> 23:46:32,912 INFO  [ChannelInterceptorBase] memberStart
>> local:org.apache.catalina.tribes.membership.MemberImpl[tcp://
>> 10.0.10.100:4000,10.0.10.100,4000, alive=272,id={-103 34 80 -19 68 -51 70
>> -91 -108 39 -84 65 50 50 103 -107 }, payload={-84 -19 0 5 115 114 0 50 111
>> ...(423)}, command={}, domain={74 79 84 84 65 95 87 65 68 ...(10)}, ]
>> notify:false peer:AS-000
>> 23:46:37,848 INFO  [DiscStore] Creating directory:
>> /usr/lib/jotta/jotta-as-prod-1.0-SNAPSHOT/var/temp/SessionStore
>> 23:46:37,930 INFO  [BasicSingletonServiceHolder] [TribesPeer [AS-000;
>> tcp://10.0.10.100:4000]] owns singleton service [PartitionManager for
>> ServiceSpace [/]]
>> 23:46:37,964 INFO  [BasicSingletonServiceHolder] [TribesPeer [AS-000;
>> tcp://10.0.10.100:4000]] resigns ownership of singleton service
>> [PartitionManager for ServiceSpace [/]]
>> 23:47:40,065 ERROR [BasicServiceRegistry] Error while starting [Holder for
>> service
>> [org.codehaus.wadi.location.partitionmanager.SimplePartitionManager@7dc2445f]
>> named [PartitionManager] in space [ServiceSpace [/]]]
>> org.codehaus.wadi.location.partitionmanager.PartitionManagerException: Partition
>> [0] is unknown.
>>        at
>> org.codehaus.wadi.location.partitionmanager.SimplePartitionManager.waitForBoot(SimplePartitionManager.java:248)
>>        at
>> org.codehaus.wadi.location.partitionmanager.SimplePartitionManager.start(SimplePartitionManager.java:119)
>>        at
>> org.codehaus.wadi.servicespace.basic.BasicServiceHolder.start(BasicServiceHolder.java:60)
>>        at
>> org.codehaus.wadi.servicespace.basic.BasicServiceRegistry.start(BasicServiceRegistry.java:152)
>>        at
>> org.codehaus.wadi.servicespace.basic.BasicServiceSpace.start(BasicServiceSpace.java:169)
>>        at
>> org.apache.geronimo.clustering.wadi.BasicWADISessionManager.doStart(BasicWADISessionManager.java:125)
>>        at
>> org.apache.geronimo.gbean.runtime.GBeanInstance.createInstance(GBeanInstance.java:948)
>>        at
>> org.apache.geronimo.gbean.runtime.GBeanInstanceState.attemptFullStart(GBeanInstanceState.java:269)
>>        at
>> org.apache.geronimo.gbean.runtime.GBeanInstanceState.start(GBeanInstanceState.java:103)
>>        at
>> org.apache.geronimo.gbean.runtime.GBeanInstanceState.startRecursive(GBeanInstanceState.java:125)
>>        at
>> org.apache.geronimo.gbean.runtime.GBeanInstance.startRecursive(GBeanInstance.java:538)
>>        at
>> org.apache.geronimo.kernel.basic.BasicKernel.startRecursiveGBean(BasicKernel.java:377)
>>        at
>> org.apache.geronimo.kernel.config.ConfigurationUtil.startConfigurationGBeans(ConfigurationUtil.java:456)
>>        at
>> org.apache.geronimo.kernel.config.ConfigurationUtil.startConfigurationGBeans(ConfigurationUtil.java:493)
>>        at
>> org.apache.geronimo.kernel.config.KernelConfigurationManager.start(KernelConfigurationManager.java:190)
>>        at
>> org.apache.geronimo.kernel.config.SimpleConfigurationManager.startConfiguration(SimpleConfigurationManager.java:546)
>>        at sun.reflect.GeneratedMethodAccessor22.invoke(Unknown Source)
>>        at
>> sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
>>        at java.lang.reflect.Method.invoke(Method.java:597)
>>        at
>> org.apache.geronimo.gbean.runtime.ReflectionMethodInvoker.invoke(ReflectionMethodInvoker.java:34)
>>        at
>> org.apache.geronimo.gbean.runtime.GBeanOperation.invoke(GBeanOperation.java:130)
>>        at
>> org.apache.geronimo.gbean.runtime.GBeanInstance.invoke(GBeanInstance.java:815)
>>        at
>> org.apache.geronimo.gbean.runtime.RawInvoker.invoke(RawInvoker.java:57)
>>        at
>> org.apache.geronimo.kernel.basic.RawOperationInvoker.invoke(RawOperationInvoker.java:35)
>>        at
>> org.apache.geronimo.kernel.basic.ProxyMethodInterceptor.intercept(ProxyMethodInterceptor.java:96)
>>        at
>> org.apache.geronimo.gbean.GBeanLifecycle$$EnhancerByCGLIB$$628b9237.startConfiguration(<generated>)
>>        at
>> org.apache.geronimo.system.main.EmbeddedDaemon.doStartup(EmbeddedDaemon.java:161)
>>        at
>> org.apache.geronimo.system.main.EmbeddedDaemon.execute(EmbeddedDaemon.java:78)
>>        at
>> org.apache.geronimo.kernel.util.MainConfigurationBootstrapper.main(MainConfigurationBootstrapper.java:45)
>>        at
>> org.apache.geronimo.cli.AbstractCLI.executeMain(AbstractCLI.java:65)
>>        at org.apache.geronimo.cli.daemon.DaemonCLI.main(DaemonCLI.java:30)
>> 23:47:40,078 ERROR [BasicWADISessionManager] Failed to stop
>> org.codehaus.wadi.servicespace.ServiceSpaceNotFoundException:
>> ServiceSpaceName not found
>>        at
>> org.codehaus.wadi.servicespace.basic.BasicServiceSpaceRegistry.unregister(BasicServiceSpaceRegistry.java:55)
>>        at
>> org.codehaus.wadi.servicespace.basic.BasicServiceSpace.unregisterServiceSpace(BasicServiceSpace.java:228)
>>        at
>> org.codehaus.wadi.servicespace.basic.BasicServiceSpace.stop(BasicServiceSpace.java:175)
>>        at
>> org.apache.geronimo.clustering.wadi.BasicWADISessionManager.doFail(BasicWADISessionManager.java:134)
>>        at
>> org.apache.geronimo.gbean.runtime.GBeanInstance.createInstance(GBeanInstance.java:978)
>>        at
>> org.apache.geronimo.gbean.runtime.GBeanInstanceState.attemptFullStart(GBeanInstanceState.java:269)
>>        at
>> org.apache.geronimo.gbean.runtime.GBeanInstanceState.start(GBeanInstanceState.java:103)
>>        at
>> org.apache.geronimo.gbean.runtime.GBeanInstanceState.startRecursive(GBeanInstanceState.java:125)
>>        at
>> org.apache.geronimo.gbean.runtime.GBeanInstance.startRecursive(GBeanInstance.java:538)
>>        at
>> org.apache.geronimo.kernel.basic.BasicKernel.startRecursiveGBean(BasicKernel.java:377)
>>        at
>> org.apache.geronimo.kernel.config.ConfigurationUtil.startConfigurationGBeans(ConfigurationUtil.java:456)
>>        at
>> org.apache.geronimo.kernel.config.ConfigurationUtil.startConfigurationGBeans(ConfigurationUtil.java:493)
>>        at
>> org.apache.geronimo.kernel.config.KernelConfigurationManager.start(KernelConfigurationManager.java:190)
>>        at
>> org.apache.geronimo.kernel.config.SimpleConfigurationManager.startConfiguration(SimpleConfigurationManager.java:546)
>>        at sun.reflect.GeneratedMethodAccessor22.invoke(Unknown Source)
>>        at
>> sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
>>        at java.lang.reflect.Method.invoke(Method.java:597)
>>        at
>> org.apache.geronimo.gbean.runtime.ReflectionMethodInvoker.invoke(ReflectionMethodInvoker.java:34)
>>        at
>> org.apache.geronimo.gbean.runtime.GBeanOperation.invoke(GBeanOperation.java:130)
>>        at
>> org.apache.geronimo.gbean.runtime.GBeanInstance.invoke(GBeanInstance.java:815)
>>        at
>> org.apache.geronimo.gbean.runtime.RawInvoker.invoke(RawInvoker.java:57)
>>        at
>> org.apache.geronimo.kernel.basic.RawOperationInvoker.invoke(RawOperationInvoker.java:35)
>>        at
>> org.apache.geronimo.kernel.basic.ProxyMethodInterceptor.intercept(ProxyMethodInterceptor.java:96)
>>        at
>> org.apache.geronimo.gbean.GBeanLifecycle$$EnhancerByCGLIB$$628b9237.startConfiguration(<generated>)
>>        at
>> org.apache.geronimo.system.main.EmbeddedDaemon.doStartup(EmbeddedDaemon.java:161)
>>        at
>> org.apache.geronimo.system.main.EmbeddedDaemon.execute(EmbeddedDaemon.java:78)
>>        at
>> org.apache.geronimo.kernel.util.MainConfigurationBootstrapper.main(MainConfigurationBootstrapper.java:45)
>>        at
>> org.apache.geronimo.cli.AbstractCLI.executeMain(AbstractCLI.java:65)
>>        at org.apache.geronimo.cli.daemon.DaemonCLI.main(DaemonCLI.java:30)
>>
>> After this failure the server stops. Over at the running instance AS-001
>> this is logged:
>>
>> 23:46:31,909 INFO  [ChannelInterceptorBase] memberAdded:tcp://{10, 0, 10,
>> 100}:4000
>> 23:46:37,929 INFO  [BasicSingletonServiceHolder] [TribesPeer [AS-001;
>> tcp://10.0.10.101:4000]] owns singleton service [PartitionManager for
>> ServiceSpace [/]]
>> 23:46:37,929 INFO  [BasicPartitionBalancerSingletonService] Queueing
>> partition rebalancing
>> 23:46:38,438 ERROR [BasicEnvelopeDispatcherManager] problem dispatching
>> message
>> java.lang.IllegalArgumentException:
>> org.codehaus.wadi.core.store.BasicStoreMotable is not a Session
>>        at
>> org.codehaus.wadi.replication.manager.basic.SessionStateHandler.newExtractFullStateExternalizable(SessionStateHandler.java:105)
>>        at
>> org.codehaus.wadi.replication.manager.basic.SessionStateHandler.extractFullState(SessionStateHandler.java:53)
>>        at
>> org.codehaus.wadi.replication.manager.basic.CreateStorageCommand.execute(CreateStorageCommand.java:45)
>>        at
>> org.codehaus.wadi.replication.manager.basic.SyncSecondaryManager.updateSecondaries(SyncSecondaryManager.java:169)
>>        at
>> org.codehaus.wadi.replication.manager.basic.SyncSecondaryManager.updateSecondaries(SyncSecondaryManager.java:114)
>>        at
>> org.codehaus.wadi.replication.manager.basic.SyncSecondaryManager.updateSecondaries(SyncSecondaryManager.java:103)
>>        at
>> org.codehaus.wadi.replication.manager.basic.SyncSecondaryManager.updateSecondariesFollowingJoiningPeer(SyncSecondaryManager.java:75)
>>        at
>> org.codehaus.wadi.replication.manager.basic.ReOrganizeSecondariesListener.receive(ReOrganizeSecondariesListener.java:53)
>>        at
>> org.codehaus.wadi.servicespace.basic.BasicServiceMonitor.notifyListeners(BasicServiceMonitor.java:124)
>>        at
>> org.codehaus.wadi.servicespace.basic.BasicServiceMonitor.processLifecycleEvent(BasicServiceMonitor.java:141)
>>        at
>> org.codehaus.wadi.servicespace.basic.BasicServiceMonitor$ServiceLifecycleEndpoint.dispatch(BasicServiceMonitor.java:148)
>>        at
>> org.codehaus.wadi.group.impl.ServiceEndpointWrapper.dispatch(ServiceEndpointWrapper.java:50)
>>        at
>> org.codehaus.wadi.group.impl.BasicEnvelopeDispatcherManager$DispatchRunner.run(BasicEnvelopeDispatcherManager.java:121)
>>        at
>> org.codehaus.wadi.servicespace.basic.BasicServiceSpaceDispatcher$ExecuteInThread.execute(BasicServiceSpaceDispatcher.java:102)
>>        at
>> org.codehaus.wadi.group.impl.BasicEnvelopeDispatcherManager.onEnvelope(BasicEnvelopeDispatcherManager.java:100)
>>        at
>> org.codehaus.wadi.group.impl.AbstractDispatcher.doOnEnvelope(AbstractDispatcher.java:104)
>>        at
>> org.codehaus.wadi.group.impl.AbstractDispatcher.onEnvelope(AbstractDispatcher.java:100)
>>        at
>> org.codehaus.wadi.servicespace.basic.ServiceSpaceEndpoint.dispatch(ServiceSpaceEndpoint.java:49)
>>        at
>> org.codehaus.wadi.group.impl.ServiceEndpointWrapper.dispatch(ServiceEndpointWrapper.java:50)
>>        at
>> org.codehaus.wadi.group.impl.BasicEnvelopeDispatcherManager$DispatchRunner.run(BasicEnvelopeDispatcherManager.java:121)
>>        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:619)
>> 23:46:40,063 INFO  [BasicPartitionBalancerSingletonService] Queueing
>> partition rebalancing
>> 23:46:42,938 WARN  [BasicPartitionBalancerSingletonService] Rebalancing
>> has failed
>> org.codehaus.wadi.group.MessageExchangeException: No correlated messages
>> received within [5000]ms
>>        at
>> org.codehaus.wadi.group.impl.AbstractDispatcher.attemptMultiRendezVous(AbstractDispatcher.java:174)
>>        at
>> org.codehaus.wadi.location.balancing.BasicPartitionBalancer.fetchBalancingInfoState(BasicPartitionBalancer.java:85)
>>        at
>> org.codehaus.wadi.location.balancing.BasicPartitionBalancer.balancePartitions(BasicPartitionBalancer.java:69)
>>        at
>> org.codehaus.wadi.location.balancing.BasicPartitionBalancerSingletonService.run(BasicPartitionBalancerSingletonService.java:85)
>>        at java.lang.Thread.run(Thread.java:619)
>> 23:46:42,939 WARN  [BasicPartitionBalancerSingletonService] Will retry
>> rebalancing in [500] ms
>> 23:46:43,439 INFO  [BasicPartitionBalancerSingletonService] Queueing
>> partition rebalancing
>> 23:47:40,269 INFO  [TcpFailureDetector] Verification complete. Member
>> disappeared[org.apache.catalina.tribes.membership.MemberImpl[tcp://{10, 0,
>> 10, 100}:4000,{10, 0, 10, 100},4000, alive=69401,id={-103 34 80 -19 68 -51
>> 70 -91 -108 39 -84 65 50 50 103 -107 }, payload={-84 -19 0 5 115 114 0 50
>> 111 ...(423)}, command={66 65 66 89 45 65 76 69 88 ...(9)}, domain={74 79 84
>> 84 65 95 87 65 68 ...(10)}, ]]
>> 23:47:40,271 INFO  [ChannelInterceptorBase] memberDisappeared:tcp://{10,
>> 0, 10, 100}:4000
>> 23:47:40,271 INFO  [BasicPartitionBalancerSingletonService] Queueing
>> partition rebalancing
>>
>> If I try to start AS-000 again the same thing happens. If we stop AS-001
>> the following is logged;
>>
>> 23:49:18,695 INFO  [SimpleStateManager] Evacuating partitions
>> 23:49:18,699 INFO  [BasicPartitionBalancerSingletonService] Queueing
>> partition rebalancing
>> 23:49:23,698 WARN  [SimpleStateManager] Partition balancer has disappeared
>> - backing off for [1000]ms
>> 23:49:24,699 INFO  [BasicPartitionBalancerSingletonService] Queueing
>> partition rebalancing
>> 23:49:29,698 WARN  [SimpleStateManager] Partition balancer has disappeared
>> - backing off for [1000]ms
>> 23:49:30,699 INFO  [BasicPartitionBalancerSingletonService] Queueing
>> partition rebalancing
>> 23:49:35,699 WARN  [SimpleStateManager] Partition balancer has disappeared
>> - backing off for [1000]ms
>> 23:49:36,700 INFO  [BasicPartitionBalancerSingletonService] Queueing
>> partition rebalancing
>> 23:49:41,699 WARN  [SimpleStateManager] Partition balancer has disappeared
>> - backing off for [1000]ms
>> 23:49:42,701 INFO  [BasicPartitionBalancerSingletonService] Queueing
>> partition rebalancing
>> 23:49:47,700 WARN  [SimpleStateManager] Partition balancer has disappeared
>> - backing off for [1000]ms
>> 23:49:48,700 INFO  [SimpleStateManager] Evacuated
>> 23:49:48,808 INFO  [AbstractExclusiveContextualiser] Unloaded
>> sessions=[36]
>> 23:49:48,843 INFO  [AbstractExclusiveContextualiser] Unloaded
>> sessions=[13]
>> 23:49:58,852 INFO  [BasicSingletonServiceHolder] [TribesPeer [AS-001;
>> tcp://10.0.10.101:4000]] resigns ownership of singleton service
>> [PartitionManager for ServiceSpace [/]]
>>
>> , however AS-001 then just hangs, and we have to kill the process to get
>> it stopped. After this we can start AS-000, update AS-001 and it always
>> seems to have no problem joining the cluster thereafter. The strange thing
>> is that this problem does not always occur, sometimes everything goes fine.
>> I can't find a consistent pattern but I've tried stopping AS-001 before
>> AS-000, and I'm sure no serializable object in the session has changed
>> between the updated and running instance.
>>
>> My gut feeling is that this is either a concurrency-related bug in WADI or
>> a network-timeout related problem. During normal operation I'm sometimes
>> seeing messages like this in the log files:
>>
>> 17:14:08,869 INFO  [TcpFailureDetector] Received
>> memberDisappeared[org.apache.catalina.tribes.membership.MemberImpl[tcp://{10,
>> 0, 10, 101}:4000,{10, 0, 10, 101},4000, alive=95659954,id={-52 -76 98 22 10
>> 71 76 -72 -122 -59 -21 -29 44 -86 38 114 }, payload={-84 -19 0 5 115 114 0
>> 50 111 ...(423)}, command={}, domain={74 79 84 84 65 95 87 65 68 ...(10)},
>> ]] message. Will verify.
>> 17:14:08,870 INFO  [TcpFailureDetector] Verification complete. Member
>> still alive[org.apache.catalina.tribes.membership.MemberImpl[tcp://{10, 0,
>> 10, 101}:4000,{10, 0, 10, 101},4000, alive=95659954,id={-52 -76 98 22 10 71
>> 76 -72 -122 -59 -21 -29 44 -86 38 114 }, payload={-84 -19 0 5 115 114 0 50
>> 111 ...(423)}, command={}, domain={74 79 84 84 65 95 87 65 68 ...(10)}, ]]
>> And lately, as traffic has increased, errors like this:
>>
>> 16:22:43,524 WARN  [UpdateReplicationCommand] Update has not been properly
>> cascaded due to a communication failure. If a targeted node has been lost,
>> state will be re-balanced automatically.
>> org.codehaus.wadi.servicespace.ServiceInvocationException:
>> org.codehaus.wadi.group.MessageExchangeException: No correlated messages
>> received within [2000]ms
>>        at
>> org.codehaus.wadi.servicespace.basic.CGLIBServiceProxyFactory$ProxyMethodInterceptor.intercept(CGLIBServiceProxyFactory.java:209)
>>        at
>> org.codehaus.wadi.replication.storage.ReplicaStorage$$EnhancerByCGLIB$$a901e91b.mergeUpdate(<generated>)
>>        at
>> org.codehaus.wadi.replication.manager.basic.UpdateReplicationCommand.cascadeUpdate(UpdateReplicationCommand.java:93)
>>        at
>> org.codehaus.wadi.replication.manager.basic.UpdateReplicationCommand.run(UpdateReplicationCommand.java:86)
>>        at
>> org.codehaus.wadi.replication.manager.basic.SyncReplicationManager.update(SyncReplicationManager.java:138)
>>        at
>> org.codehaus.wadi.replication.manager.basic.LoggingReplicationManager.update(LoggingReplicationManager.java:100)
>>        at
>> org.codehaus.wadi.core.session.AbstractReplicableSession.onEndProcessing(AbstractReplicableSession.java:49)
>>        at
>> org.codehaus.wadi.core.session.AtomicallyReplicableSession.onEndProcessing(AtomicallyReplicableSession.java:58)
>>        at
>> org.apache.geronimo.clustering.wadi.WADISessionAdaptor.onEndAccess(WADISessionAdaptor.java:77)
>>        at
>> org.apache.geronimo.jetty6.cluster.ClusteredSessionManager.complete(ClusteredSessionManager.java:60)
>>        at
>> org.mortbay.jetty.servlet.SessionHandler.handle(SessionHandler.java:198)
>>        at
>> org.apache.geronimo.jetty6.cluster.ClusteredSessionHandler.doHandle(ClusteredSessionHandler.java:59)
>>        at
>> org.apache.geronimo.jetty6.cluster.ClusteredSessionHandler$ActualHandler.handle(ClusteredSessionHandler.java:66)
>>        at
>> org.apache.geronimo.jetty6.cluster.AbstractClusteredPreHandler$WebClusteredInvocation.invokeLocally(AbstractClusteredPreHandler.java:71)
>>        at
>> org.apache.geronimo.jetty6.cluster.wadi.WADIClusteredPreHandler$WADIWebClusteredInvocation.access$000(WADIClusteredPreHandler.java:52)
>>        at
>> org.apache.geronimo.jetty6.cluster.wadi.WADIClusteredPreHandler$WADIWebClusteredInvocation$1.doFilter(WADIClusteredPreHandler.java:64)
>>        at
>> org.codehaus.wadi.web.impl.WebInvocation.invoke(WebInvocation.java:116)
>>        at
>> org.codehaus.wadi.core.contextualiser.MemoryContextualiser.handleLocally(MemoryContextualiser.java:71)
>>        at
>> org.codehaus.wadi.core.contextualiser.AbstractExclusiveContextualiser.handle(AbstractExclusiveContextualiser.java:94)
>>        at
>> org.codehaus.wadi.core.contextualiser.AbstractMotingContextualiser.contextualise(AbstractMotingContextualiser.java:37)
>>        at
>> org.codehaus.wadi.core.manager.StandardManager.processStateful(StandardManager.java:150)
>>        at
>> org.codehaus.wadi.core.manager.StandardManager.contextualise(StandardManager.java:142)
>>        at
>> org.codehaus.wadi.core.manager.ClusteredManager.contextualise(ClusteredManager.java:81)
>>        at
>> org.apache.geronimo.jetty6.cluster.wadi.WADIClusteredPreHandler$WADIWebClusteredInvocation.invoke(WADIClusteredPreHandler.java:72)
>>        at
>> org.apache.geronimo.jetty6.cluster.AbstractClusteredPreHandler.handle(AbstractClusteredPreHandler.java:39)
>>        at
>> org.apache.geronimo.jetty6.cluster.ClusteredSessionHandler.handle(ClusteredSessionHandler.java:51)
>>        at
>> org.mortbay.jetty.handler.ContextHandler.handle(ContextHandler.java:765)
>>        at
>> org.mortbay.jetty.webapp.WebAppContext.handle(WebAppContext.java:417)
>>        at
>> org.apache.geronimo.jetty6.handler.TwistyWebAppContext.access$101(TwistyWebAppContext.java:41)
>>        at
>> org.apache.geronimo.jetty6.handler.TwistyWebAppContext$TwistyHandler.handle(TwistyWebAppContext.java:66)
>>        at
>> org.apache.geronimo.jetty6.handler.ThreadClassloaderHandler.handle(ThreadClassloaderHandler.java:46)
>>        at
>> org.apache.geronimo.jetty6.handler.InstanceContextHandler.handle(InstanceContextHandler.java:58)
>>        at
>> org.apache.geronimo.jetty6.handler.UserTransactionHandler.handle(UserTransactionHandler.java:48)
>>        at
>> org.apache.geronimo.jetty6.handler.ComponentContextHandler.handle(ComponentContextHandler.java:47)
>>        at
>> org.apache.geronimo.jetty6.handler.TwistyWebAppContext.handle(TwistyWebAppContext.java:60)
>>        at
>> org.mortbay.jetty.handler.ContextHandlerCollection.handle(ContextHandlerCollection.java:230)
>>        at
>> org.mortbay.jetty.handler.HandlerCollection.handle(HandlerCollection.java:114)
>>        at
>> org.mortbay.jetty.handler.HandlerWrapper.handle(HandlerWrapper.java:152)
>>        at org.mortbay.jetty.Server.handle(Server.java:326)
>>        at
>> org.mortbay.jetty.HttpConnection.handleRequest(HttpConnection.java:534)
>>        at
>> org.mortbay.jetty.HttpConnection$RequestHandler.content(HttpConnection.java:879)
>>        at org.mortbay.jetty.HttpParser.parseNext(HttpParser.java:747)
>>        at org.mortbay.jetty.HttpParser.parseAvailable(HttpParser.java:218)
>>        at org.mortbay.jetty.HttpConnection.handle(HttpConnection.java:404)
>>        at
>> org.mortbay.io.nio.SelectChannelEndPoint.run(SelectChannelEndPoint.java:409)
>>        at org.apache.geronimo.pool.ThreadPool$1.run(ThreadPool.java:214)
>>        at
>> org.apache.geronimo.pool.ThreadPool$ContextClassLoaderRunnable.run(ThreadPool.java:344)
>>        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:619)
>> Caused by: org.codehaus.wadi.group.MessageExchangeException: No correlated
>> messages received within [2000]ms
>>        at
>> org.codehaus.wadi.group.impl.AbstractDispatcher.attemptMultiRendezVous(AbstractDispatcher.java:174)
>>        at
>> org.codehaus.wadi.servicespace.basic.BasicServiceInvoker.invokeOnPeers(BasicServiceInvoker.java:90)
>>        at
>> org.codehaus.wadi.servicespace.basic.BasicServiceInvoker.invoke(BasicServiceInvoker.java:69)
>>        at
>> org.codehaus.wadi.servicespace.basic.CGLIBServiceProxyFactory$ProxyMethodInterceptor.intercept(CGLIBServiceProxyFactory.java:193)
>>        ... 49 more
>>
>> Does anyone have some insight into what might be causing this, or where
>> the timeouts can be increased if there are any?
>>
>> I'm thinking that a static WADI configuration can be more stable than the
>> dynamic setup we have not which relies upon multicasting. Does anyone have
>> experience with similar setups?
>>
>> Thanks!
>>
>> Trygve Hardersen - Jotta
>>
>>
>>
>>
>>
>

Mime
View raw message