[jboss-jira] [JBoss JIRA] (WFLY-5680) InfinispanBatchContext trying to resume committed tx

Paul Ferraro (JIRA) issues at jboss.org
Thu Nov 19 08:01:02 EST 2015


     [ https://issues.jboss.org/browse/WFLY-5680?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Paul Ferraro updated WFLY-5680:
-------------------------------
    Summary: InfinispanBatchContext trying to resume committed tx  (was: InfinispanBatchContext trying to resume commited txn)


> InfinispanBatchContext trying to resume committed tx
> ----------------------------------------------------
>
>                 Key: WFLY-5680
>                 URL: https://issues.jboss.org/browse/WFLY-5680
>             Project: WildFly
>          Issue Type: Bug
>          Components: Clustering
>    Affects Versions: 10.0.0.CR4
>         Environment: Linux x86_64
>            Reporter: Richard Achmatowicz
>            Assignee: Paul Ferraro
>            Priority: Critical
>         Attachments: resume-commited-txn-output.txt
>
>
> When running the test case RemoteFailoverTestCase.testStatefulFailover, the test hangs due to an exception occurring. The exception is the following:
> {noformat}
> [31m11:24:05,814 ERROR [org.infinispan.interceptors.InvocationContextInterceptor] (EJB default - 3) ISPN000136: Execution error: java.lang.IllegalStateException: Transaction DummyTransaction{xid=DummyXid{id=47}, status=3} is not in a valid state to be invoking cache operations on.
> {noformat} 
> This exception arose some time ago and was linked to incorrect suspend/resume of threads by the InfinispanBatching mechanism. This is just a suggestion as to what might be the root cause.
> Attached is a log of the test case run with TRACE logging on DummyTransactionManager which coordinates the suspends and resumes. I have also performed a stack trace dump on every call to suspend and resume to highlight the origin of the call. 



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


More information about the jboss-jira mailing list