[
https://issues.apache.org/jira/browse/YARN-3878?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14613049#comment-14613049
]
Devaraj K commented on YARN-3878:
---------------------------------
bq. Although as currently isdrained is not checked anywhere except in
DrainDispatcher#await, we can easily remove it. And if somebody else requires,
they can add it later. Thoughts ?
eventQueue is getting passed from DrainDispatcher constructers and the same can
be used in DrainDispatcher.await(). If anybody else wants it in future then
they can add the method based on the scenario. I don't see any real reason for
keeping this method.
bq. In this case, close() directly calls stop() so both of them are in essence
same same. Any reason you prefer close over stop ?
I agree both do the same but we can avoid the javac warning for disp i.e
{{Resource leak: 'disp' is never closed}}.
And also there are some rawtypes warnings showing up in the test, you can
probably suppress them as well.
> AsyncDispatcher can hang while stopping if it is configured for draining
> events on stop
> ---------------------------------------------------------------------------------------
>
> Key: YARN-3878
> URL: https://issues.apache.org/jira/browse/YARN-3878
> Project: Hadoop YARN
> Issue Type: Bug
> Affects Versions: 2.7.0
> Reporter: Varun Saxena
> Assignee: Varun Saxena
> Priority: Critical
> Attachments: YARN-3878.01.patch, YARN-3878.02.patch
>
>
> The sequence of events is as under :
> # RM is stopped while putting a RMStateStore Event to RMStateStore's
> AsyncDispatcher. This leads to an Interrupted Exception being thrown.
> # As RM is being stopped, RMStateStore's AsyncDispatcher is also stopped. On
> {{serviceStop}}, we will check if all events have been drained and wait for
> event queue to drain(as RM State Store dispatcher is configured for queue to
> drain on stop).
> # This condition never becomes true and AsyncDispatcher keeps on waiting
> incessantly for dispatcher event queue to drain till JVM exits.
> *Initial exception while posting RM State store event to queue*
> {noformat}
> 2015-06-27 20:08:35,922 DEBUG [main] service.AbstractService
> (AbstractService.java:enterState(452)) - Service: Dispatcher entered state
> STOPPED
> 2015-06-27 20:08:35,923 WARN [AsyncDispatcher event handler]
> event.AsyncDispatcher (AsyncDispatcher.java:handle(247)) - AsyncDispatcher
> thread interrupted
> java.lang.InterruptedException
> at
> java.util.concurrent.locks.AbstractQueuedSynchronizer.acquireInterruptibly(AbstractQueuedSynchronizer.java:1219)
> at
> java.util.concurrent.locks.ReentrantLock.lockInterruptibly(ReentrantLock.java:340)
> at
> java.util.concurrent.LinkedBlockingQueue.put(LinkedBlockingQueue.java:338)
> at
> org.apache.hadoop.yarn.event.AsyncDispatcher$GenericEventHandler.handle(AsyncDispatcher.java:244)
> at
> org.apache.hadoop.yarn.server.resourcemanager.recovery.RMStateStore.updateApplicationAttemptState(RMStateStore.java:652)
> at
> org.apache.hadoop.yarn.server.resourcemanager.rmapp.attempt.RMAppAttemptImpl.rememberTargetTransitionsAndStoreState(RMAppAttemptImpl.java:1173)
> at
> org.apache.hadoop.yarn.server.resourcemanager.rmapp.attempt.RMAppAttemptImpl.access$3300(RMAppAttemptImpl.java:109)
> at
> org.apache.hadoop.yarn.server.resourcemanager.rmapp.attempt.RMAppAttemptImpl$ContainerFinishedTransition.transition(RMAppAttemptImpl.java:1650)
> at
> org.apache.hadoop.yarn.server.resourcemanager.rmapp.attempt.RMAppAttemptImpl$ContainerFinishedTransition.transition(RMAppAttemptImpl.java:1619)
> at
> org.apache.hadoop.yarn.state.StateMachineFactory$MultipleInternalArc.doTransition(StateMachineFactory.java:385)
> at
> org.apache.hadoop.yarn.state.StateMachineFactory.doTransition(StateMachineFactory.java:302)
> at
> org.apache.hadoop.yarn.state.StateMachineFactory.access$300(StateMachineFactory.java:46)
> at
> org.apache.hadoop.yarn.state.StateMachineFactory$InternalStateMachine.doTransition(StateMachineFactory.java:448)
> at
> org.apache.hadoop.yarn.server.resourcemanager.rmapp.attempt.RMAppAttemptImpl.handle(RMAppAttemptImpl.java:786)
> at
> org.apache.hadoop.yarn.server.resourcemanager.rmapp.attempt.RMAppAttemptImpl.handle(RMAppAttemptImpl.java:108)
> at
> org.apache.hadoop.yarn.server.resourcemanager.ResourceManager$ApplicationAttemptEventDispatcher.handle(ResourceManager.java:838)
> {noformat}
> *JStack of AsyncDispatcher hanging on stop*
> {noformat}
> "AsyncDispatcher event handler" prio=10 tid=0x00007fb980222800 nid=0x4b1e
> waiting on condition [0x00007fb9654e9000]
> java.lang.Thread.State: WAITING (parking)
> at sun.misc.Unsafe.park(Native Method)
> - parking to wait for <0x0000000700b79250> (a
> java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject)
> at java.util.concurrent.locks.LockSupport.park(LockSupport.java:186)
> at
> java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject.await(AbstractQueuedSynchronizer.java:2043)
> at
> java.util.concurrent.LinkedBlockingQueue.take(LinkedBlockingQueue.java:442)
> at
> org.apache.hadoop.yarn.event.AsyncDispatcher$1.run(AsyncDispatcher.java:113)
> at java.lang.Thread.run(Thread.java:744)
> "main" prio=10 tid=0x00007fb98000a800 nid=0x49c3 in Object.wait()
> [0x00007fb989851000]
> java.lang.Thread.State: TIMED_WAITING (on object monitor)
> at java.lang.Object.wait(Native Method)
> - waiting on <0x0000000700b79430> (a java.lang.Object)
> at
> org.apache.hadoop.yarn.event.AsyncDispatcher.serviceStop(AsyncDispatcher.java:156)
> - locked <0x0000000700b79430> (a java.lang.Object)
> at
> org.apache.hadoop.service.AbstractService.stop(AbstractService.java:221)
> - locked <0x0000000700b79420> (a java.lang.Object)
> at
> org.apache.hadoop.yarn.server.resourcemanager.recovery.RMStateStore.serviceStop(RMStateStore.java:515)
> at
> org.apache.hadoop.service.AbstractService.stop(AbstractService.java:221)
> - locked <0x0000000700b79630> (a java.lang.Object)
> at
> org.apache.hadoop.service.AbstractService.close(AbstractService.java:250)
> at
> org.apache.hadoop.yarn.server.resourcemanager.ResourceManager$RMActiveServices.serviceStop(ResourceManager.java:599)
> {noformat}
> We keep on getting below logs
> {noformat}
> 2015-06-27 20:08:35,926 INFO [main] event.AsyncDispatcher
> (AsyncDispatcher.java:serviceStop(140)) - AsyncDispatcher is draining to
> stop, igonring any new events.
> 2015-06-27 20:08:36,926 INFO [main] event.AsyncDispatcher
> (AsyncDispatcher.java:serviceStop(144)) - Waiting for AsyncDispatcher to
> drain. Thread state is :WAITING
> 2015-06-27 20:08:37,927 INFO [main] event.AsyncDispatcher
> (AsyncDispatcher.java:serviceStop(144)) - Waiting for AsyncDispatcher to
> drain. Thread state is :WAITING
> 2015-06-27 20:08:38,927 INFO [main] event.AsyncDispatcher
> (AsyncDispatcher.java:serviceStop(144)) - Waiting for AsyncDispatcher to
> drain. Thread state is :WAITING
> 2015-06-27 20:08:39,928 INFO [main] event.AsyncDispatcher
> (AsyncDispatcher.java:serviceStop(144)) - Waiting for AsyncDispatcher to
> drain. Thread state is :WAITING
> 2015-06-27 20:08:40,929 INFO [main] event.AsyncDispatcher
> (AsyncDispatcher.java:serviceStop(144)) - Waiting for AsyncDispatcher to
> drain. Thread state is :WAITING
> 2015-06-27 20:08:41,929 INFO [main] event.AsyncDispatcher
> (AsyncDispatcher.java:serviceStop(144)) - Waiting for AsyncDispatcher to
> drain. Thread state is :WAITING
> 2015-06-27 20:08:42,930 INFO [main] event.AsyncDispatcher
> (AsyncDispatcher.java:serviceStop(144)) - Waiting for AsyncDispatcher to
> drain. Thread state is :WAITING
> 2015-06-27 20:08:43,930 INFO [main] event.AsyncDispatcher
> (AsyncDispatcher.java:serviceStop(144)) - Waiting for AsyncDispatcher to
> drain. Thread state is :WAITING
> 2015-06-27 20:08:44,931 INFO [main] event.AsyncDispatcher
> (AsyncDispatcher.java:serviceStop(144)) - Waiting for AsyncDispatcher to
> drain. Thread state is :WAITING
> 2015-06-27 20:08:45,931 INFO [main] event.AsyncDispatcher
> (AsyncDispatcher.java:serviceStop(144)) - Waiting for AsyncDispatcher to
> drain. Thread state is :WAITING
> 2015-06-27 20:08:46,932 INFO [main] event.AsyncDispatcher
> (AsyncDispatcher.java:serviceStop(144)) - Waiting for AsyncDispatcher to
> drain. Thread state is :WAITING
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
