[jboss-jira] [JBoss JIRA] (WFLY-6319) NPE in LocalEjbReceiver.removeClusterNodes during server shutdown

[Richard Achmatowicz (JIRA)]

    [ https://issues.jboss.org/browse/WFLY-6319?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13179178#comment-13179178 ] 

Richard Achmatowicz edited comment on WFLY-6319 at 3/18/16 11:43 AM:
---------------------------------------------------------------------

Hi Dimitii

I've been trying to find out why this NPE is occurring in the first place. It can certainly happen if the EJBClientContext which holds the cluster contexts being removed has already been closed (any request to get a cluster context returns null), but i'm not sure if or how that can happen yet.

I suspect it is related to the fact that we do not clean up stale cluster contexts in EJBClientContexts when a cluster is entirely removed in LocalEJBReceiver.registryRemoved(). I think this may be the source of the problem and should probably be fixed as it was in VersionOneProtocolChannelReceiver.

But its only a guess and needs to be confirmed.


was (Author: rachmato):
Hi Dimitii

I've been trying to find out why this NPE is occurring in the first place. It can certainly happen if the EJBClientContext which holds the cluster contexts being removed has already been closed (any request to get a cluster context returns null), but i'm not sure if or how that can happen yet.

I suspect it is related to the fact that we do not clean up stale cluster contexts in EJBClientContexts when a cluster is entirely removed in LocalEJBReceiver.registryRemoved(). I think this may be the source of the problem and should probably be fixed as it was in VersionOneProtocolChannelReceiver.

> NPE in LocalEjbReceiver.removeClusterNodes during server shutdown
> -----------------------------------------------------------------
>
>                 Key: WFLY-6319
>                 URL: https://issues.jboss.org/browse/WFLY-6319
>             Project: WildFly
>          Issue Type: Bug
>          Components: Clustering, EJB
>            Reporter: Ladislav Thon
>            Assignee: Dmitrii Tikhomirov
>
> In a failover test that passivates sessions to a SQL database (though I think this isn't really relevant much), I've seen a NPE in {{LocalEjbReceiver.removeClusterNodes}}:
> {code}
> 04:09:29,330 WARN  [org.jboss.msc.service.fail] (MSC service thread 1-6) MSC000004: Failure during stop of service jboss.ejb3.localEjbReceiver.value: java.lang.NullPointerException
> 	at org.jboss.as.ejb3.remote.LocalEjbReceiver.removeClusterNodes(LocalEjbReceiver.java:701)
> 	at org.jboss.as.ejb3.remote.LocalEjbReceiver.disassociate(LocalEjbReceiver.java:681)
> 	at org.jboss.ejb.client.EJBClientContext.unregisterEJBReceiver(EJBClientContext.java:446)
> 	at org.jboss.ejb.client.EJBReceiverContext.close(EJBReceiverContext.java:57)
> 	at org.jboss.as.ejb3.remote.LocalEjbReceiver.stop(LocalEjbReceiver.java:410)
> 	at org.jboss.msc.service.ServiceControllerImpl$StopTask.stopService(ServiceControllerImpl.java:2056)
> 	at org.jboss.msc.service.ServiceControllerImpl$StopTask.run(ServiceControllerImpl.java:2017)
> 	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
> 	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
> 	at java.lang.Thread.run(Thread.java:745)
> {code}
> This happens during server shutdown, so it probably doesn't affect anything.
> Logs:
> https://jenkins.mw.lab.eng.bos.redhat.com/hudson/job/eap-7x-failover-db-session-shutdown-repl-sync-mysql/42/console-perf21/
> https://jenkins.mw.lab.eng.bos.redhat.com/hudson/job/eap-7x-failover-db-session-shutdown-repl-sync-postgresql/7/console-perf20/



--
This message was sent by Atlassian JIRA
(v6.4.11#64026)