[jboss-jira] [JBoss JIRA] (WFLY-5836) Stale session data received when using REPL SYNC on jvmkill

[Radoslav Husar (JIRA)]

     [ https://issues.jboss.org/browse/WFLY-5836?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Radoslav Husar closed WFLY-5836.
--------------------------------
      Assignee: Radoslav Husar  (was: Paul Ferraro)
    Resolution: Rejected


This can happen in this scenario, when the server crashes after the client response is committed but before the transaction was committed. An indication of a problem would have been "off by 2" versions of the session implying a stale data read.

> Stale session data received when using REPL SYNC on jvmkill
> -----------------------------------------------------------
>
>                 Key: WFLY-5836
>                 URL: https://issues.jboss.org/browse/WFLY-5836
>             Project: WildFly
>          Issue Type: Bug
>          Components: Clustering
>    Affects Versions: 10.0.0.CR4
>            Reporter: Michal Vinkler
>            Assignee: Radoslav Husar
>            Priority: Minor
>
> Client received older data when using REPL SYNC right after failing perf21 (jvmkill).
> -- seen only once, hardly reproducible
> Timeline & stacktrace:
> {code}
> 2015/11/27 15:05:40:255 EST [INFO ][TestController] HOST perf17.mw.lab.eng.bos.redhat.com:rootProcess:c - Failing node 3 (perf21)
> 2015/11/27 15:05:40:459 EST [INFO ][TestController] HOST perf17.mw.lab.eng.bos.redhat.com:rootProcess:c - Node 3 (perf21) is down.
> 2015/11/27 15:05:40:459 EST [INFO ][TestController] HOST perf17.mw.lab.eng.bos.redhat.com:rootProcess:c - Cluster state: [UP, UP, UP, DOWN].
> 2015/11/27 15:05:45:758 EST [INFO ][Runner - 725] HOST perf17.mw.lab.eng.bos.redhat.com:rootProcess:c - Failover detected, JvmRoute changed. perf21 -> perf20
> 2015/11/27 15:05:45:774 EST [INFO ][Runner - 1962] HOST perf17.mw.lab.eng.bos.redhat.com:rootProcess:c - Failover detected, JvmRoute changed. perf21 -> perf19
> :
> :
> 2015/11/27 15:05:49:999 EST [INFO ][Runner - 522] HOST perf17.mw.lab.eng.bos.redhat.com:rootProcess:c - Failover detected, JvmRoute changed. perf21 -> perf20
> 2015/11/27 15:05:50:533 EST [INFO ][Runner - 314] HOST perf17.mw.lab.eng.bos.redhat.com:rootProcess:c - Failover detected, JvmRoute changed. perf21 -> perf20
> 2015/11/27 15:05:53:908 EST [WARN ][Runner - 14] HOST perf17.mw.lab.eng.bos.redhat.com:rootProcess:c - Error sampling data:  <org.jboss.smartfrog.loaddriver.RequestProcessingException: Stale session data received. Expected 139, received 138, Runner: 14>
>         org.jboss.smartfrog.loaddriver.RequestProcessingException: Stale session data received. Expected 139, received 138, Runner: 14
>         at org.jboss.smartfrog.loaddriver.http.AbstractSerialNumberValidatorFactoryImpl$SerialNumberValidator.processRequest(AbstractSerialNumberValidatorFactoryImpl.java:133)
>         at org.jboss.smartfrog.loaddriver.CompoundRequestProcessorFactoryImpl$CompoundRequestProcessor.processRequest(CompoundRequestProcessorFactoryImpl.java:52)
>         at org.jboss.smartfrog.loaddriver.Runner.run(Runner.java:103)
>         at java.lang.Thread.run(Thread.java:745)
> {code}
> Link
> http://jenkins.mw.lab.eng.bos.redhat.com/hudson/job/eap-7x-failover-http-session-jvmkill-repl-sync/11/console-perf17/



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