Mailing-List: contact dev-help@geode.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@geode.apache.org
Date: Thu, 27 Apr 2017 06:28:04 +0000 (UTC)
From: "ASF subversion and git services (JIRA)" <jira@apache.org>
To: dev@geode.apache.org
Message-ID: <JIRA.13065945.1492795553000.63233.1493274484681@Atlassian.JIRA>
In-Reply-To: <JIRA.13065945.1492795553000@Atlassian.JIRA>
References: <JIRA.13065945.1492795553000@Atlassian.JIRA> <JIRA.13065945.1492795553175@jira-lw-us.apache.org>
Subject: [jira] [Commented] (GEODE-2808) Deadlock in tomcat session module
 during session expiration
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Thu, 27 Apr 2017 06:28:11 -0000


    [ https://issues.apache.org/jira/browse/GEODE-2808?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15986091#comment-15986091 ] 

ASF subversion and git services commented on GEODE-2808:
--------------------------------------------------------

Commit 80a95f6047ad64d165744f5ae4088b409484d50f in geode's branch refs/heads/feature/GEM-1299 from [~upthewaterspout]
[ https://git-wip-us.apache.org/repos/asf?p=geode.git;h=80a95f6 ]

GEODE-2808 - Fixing lock ordering issues in DeltaSession

Region expiration of sessions and explicit expiration of sessions had
lock ordering issues. Fixing the code so that expiration goes through
the region entry lock first, before getting the lock on StandardSession.

Adding a workaround for the fact that liferay calls removeAttribute
from within session expiration by ignoreing remoteAttribute calls during
expiration.

This closes #472


> Deadlock in tomcat session module during session expiration
> -----------------------------------------------------------
>
>                 Key: GEODE-2808
>                 URL: https://issues.apache.org/jira/browse/GEODE-2808
>             Project: Geode
>          Issue Type: Bug
>          Components: http session
>            Reporter: Dan Smith
>            Assignee: Dan Smith
>             Fix For: 1.2.0
>
>
> We observed a deadlock in the tomcat session state replication module due to the way we do expiration. Here are the threads that are involved:
> {noformat}
> Found one Java-level deadlock:
> =============================
> "http-nio-21064-exec-2":
>   waiting to lock monitor 0x00007f9a04bde298 (object 0x00000007947da1f0, a org.apache.geode.internal.cache.VersionedStatsLRURegionEntryHeapObjectKey),
>   which is held by "Timer-1"
> "Timer-1":
>   waiting to lock monitor 0x00007f99a4004e38 (object 0x00000007947da240, a org.apache.geode.modules.session.catalina.DeltaSession8),
>   which is held by "http-nio-21064-exec-2"
> Java stack information for the threads listed above:
> ===================================================
> "http-nio-21064-exec-2":
> 	at org.apache.geode.internal.cache.AbstractRegionMap.destroy(AbstractRegionMap.java:1228)
> 	- waiting to lock <0x00000007947da1f0> (a org.apache.geode.internal.cache.VersionedStatsLRURegionEntryHeapObjectKey)
> 	at org.apache.geode.internal.cache.LocalRegion.mapDestroy(LocalRegion.java:6785)
> 	at org.apache.geode.internal.cache.LocalRegion.mapDestroy(LocalRegion.java:6762)
> 	at org.apache.geode.internal.cache.LocalRegionDataView.destroyExistingEntry(LocalRegionDataView.java:55)
> 	at org.apache.geode.internal.cache.LocalRegion.basicDestroy(LocalRegion.java:6724)
> 	at org.apache.geode.internal.cache.LocalRegion.validatedDestroy(LocalRegion.java:1110)
> 	at org.apache.geode.internal.cache.LocalRegion.destroy(LocalRegion.java:1095)
> 	at org.apache.geode.internal.cache.AbstractRegion.destroy(AbstractRegion.java:271)
> 	at org.apache.geode.modules.session.catalina.AbstractSessionCache.destroySession(AbstractSessionCache.java:72)
> 	at org.apache.geode.modules.session.catalina.DeltaSessionManager.remove(DeltaSessionManager.java:405)
> 	at org.apache.catalina.session.StandardSession.expire(StandardSession.java:850)
> 	- locked <0x00000007947da240> (a org.apache.geode.modules.session.catalina.DeltaSession8)
> 	at org.apache.catalina.session.StandardSession.isValid(StandardSession.java:682)
> 	at org.apache.catalina.connector.Request.doGetSession(Request.java:2917)
> 	at org.apache.catalina.connector.Request.getSession(Request.java:2367)
> 	at org.apache.geode.modules.session.catalina.CommitSessionValve.invoke(CommitSessionValve.java:50)
> 	at org.apache.geode.modules.session.catalina.JvmRouteBinderValve.invoke(JvmRouteBinderValve.java:46)
> 	at org.apache.catalina.authenticator.AuthenticatorBase.invoke(AuthenticatorBase.java:474)
> 	at org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:140)
> 	at org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:79)
> 	at org.apache.catalina.core.StandardEngineValve.invoke(StandardEngineValve.java:87)
> 	at org.apache.geode.modules.session.catalina.JvmRouteBinderValve.invoke(JvmRouteBinderValve.java:46)
> 	at org.apache.catalina.connector.CoyoteAdapter.service(CoyoteAdapter.java:349)
> 	at org.apache.coyote.http11.Http11Processor.service(Http11Processor.java:783)
> 	at org.apache.coyote.AbstractProcessorLight.process(AbstractProcessorLight.java:66)
> 	at org.apache.coyote.AbstractProtocol$ConnectionHandler.process(AbstractProtocol.java:789)
> 	at org.apache.tomcat.util.net.NioEndpoint$SocketProcessor.doRun(NioEndpoint.java:1437)
> 	at org.apache.tomcat.util.net.SocketProcessorBase.run(SocketProcessorBase.java:49)
> 	- locked <0x00000007948d6870> (a org.apache.tomcat.util.net.NioEndpoint$NioSocketWrapper)
> 	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
> 	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
> 	at org.apache.tomcat.util.threads.TaskThread$WrappingRunnable.run(TaskThread.java:61)
> 	at java.lang.Thread.run(Thread.java:745)
> "Timer-1":
> 	at org.apache.catalina.session.StandardSession.expire(StandardSession.java:791)
> 	- waiting to lock <0x00000007947da240> (a org.apache.geode.modules.session.catalina.DeltaSession8)
> 	at org.apache.catalina.session.StandardSession.expire(StandardSession.java:766)
> 	at org.apache.geode.modules.session.catalina.DeltaSession8.processExpired(DeltaSession8.java:320)
> 	at org.apache.geode.modules.session.catalina.callback.SessionExpirationCacheListener.afterDestroy(SessionExpirationCacheListener.java:56)
> 	at org.apache.geode.internal.cache.EnumListenerEvent$AFTER_DESTROY.dispatchEvent(EnumListenerEvent.java:151)
> 	at org.apache.geode.internal.cache.LocalRegion.dispatchEvent(LocalRegion.java:8751)
> 	at org.apache.geode.internal.cache.LocalRegion.dispatchListenerEvent(LocalRegion.java:7245)
> 	at org.apache.geode.internal.cache.LocalRegion.invokeDestroyCallbacks(LocalRegion.java:7052)
> 	at org.apache.geode.internal.cache.EntryEventImpl.invokeCallbacks(EntryEventImpl.java:2293)
> 	at org.apache.geode.internal.cache.AbstractRegionEntry.dispatchListenerEvents(AbstractRegionEntry.java:140)
> 	at org.apache.geode.internal.cache.LocalRegion.basicDestroyPart2(LocalRegion.java:6992)
> 	at org.apache.geode.internal.cache.AbstractRegionMap.destroy(AbstractRegionMap.java:1419)
> 	- locked <0x00000007947da1f0> (a org.apache.geode.internal.cache.VersionedStatsLRURegionEntryHeapObjectKey)
> 	at org.apache.geode.internal.cache.LocalRegion.mapDestroy(LocalRegion.java:6785)
> 	at org.apache.geode.internal.cache.LocalRegion.mapDestroy(LocalRegion.java:6762)
> 	at org.apache.geode.internal.cache.LocalRegionDataView.destroyExistingEntry(LocalRegionDataView.java:55)
> 	at org.apache.geode.internal.cache.LocalRegion.basicDestroy(LocalRegion.java:6724)
> 	at org.apache.geode.internal.cache.LocalRegion.expireDestroy(LocalRegion.java:6864)
> 	at org.apache.geode.internal.cache.EntryExpiryTask.destroy(EntryExpiryTask.java:119)
> 	at org.apache.geode.internal.cache.ExpiryTask.expire(ExpiryTask.java:278)
> 	at org.apache.geode.internal.cache.ExpiryTask.expire(ExpiryTask.java:232)
> 	at org.apache.geode.internal.cache.EntryExpiryTask.basicPerformTimeout(EntryExpiryTask.java:305)
> 	at org.apache.geode.internal.cache.LocalRegion.performExpiryTimeout(LocalRegion.java:801)
> 	at org.apache.geode.internal.cache.EntryExpiryTask.performTimeout(EntryExpiryTask.java:219)
> 	at org.apache.geode.internal.cache.ExpiryTask.runInThreadPool(ExpiryTask.java:370)
> 	at org.apache.geode.internal.cache.ExpiryTask.run2(ExpiryTask.java:318)
> 	at org.apache.geode.internal.SystemTimer$SystemTimerTask.run(SystemTimer.java:444)
> 	at java.util.TimerThread.mainLoop(Timer.java:555)
> 	at java.util.TimerThread.run(Timer.java:505)
> Found 1 deadlock.
> {noformat}
> The basic issue is that we configure expiration on the geode region to expire sessions, and the tomcat container will also expire sessions when the get to old. But those two threads get locks in a different order:
> Tomcat expiration (http-nio-21064-exec-2)
>   # sync session in the tomcat code (StandardSession.expire)
>   # sync region entry in the geode code (AbstractRegionMap.destroy)
> region expiration
>   # sync region entry in the geode code (AbstractRegionMap.destroy
>   # sync session in geode code (StandardSession.expire)


--
This message was sent by Atlassian JIRA
(v6.3.15#6346)