[
https://issues.apache.org/jira/browse/YARN-1965?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14603103#comment-14603103
]
Mit Desai commented on YARN-1965:
---------------------------------
Overall patch Looks good. Few minor nits.
* There should be a space between () and { here
{{public static final ExecutorService getClientExecutor(){}}
* In testStandAloneClient(), we need spaces near the brackets. Change
{{}finally{}} to {{} finally {}}
* In testConnectionIdleTimeouts(), we need space near the brackets. Change
{{}finally{}} to {{} finally {}}
* testInterrupted needs to be indented.
* In doErrorTest, testRTEDuringConnectionSetup stopping the client before the
server makes more sense. Swap the stop calls in finally block
* In testSocketFactoryException,testIpcConnectTimeout {{client.stop()}} should
be within finally block
* Is there a need to move {{Client client = new Client(LongWritable.class,
conf, spyFactory);}} in testRTEDuringConnectionSetup?
> Interrupted exception when closing YarnClient
> ---------------------------------------------
>
> Key: YARN-1965
> URL: https://issues.apache.org/jira/browse/YARN-1965
> Project: Hadoop YARN
> Issue Type: Bug
> Components: api
> Affects Versions: 2.3.0
> Reporter: Oleg Zhurakousky
> Assignee: Kuhu Shukla
> Priority: Minor
> Labels: newbie
> Attachments: YARN-1965-v2.patch, YARN-1965.patch
>
>
> Its more of a nuisance then a bug, but nevertheless
> {code}
> 16:16:48,709 ERROR pool-1-thread-1 ipc.Client:195 - Interrupted while waiting
> for clientExecutorto stop
> java.lang.InterruptedException
> at
> java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject.awaitNanos(AbstractQueuedSynchronizer.java:2072)
> at
> java.util.concurrent.ThreadPoolExecutor.awaitTermination(ThreadPoolExecutor.java:1468)
> at
> org.apache.hadoop.ipc.Client$ClientExecutorServiceFactory.unrefAndCleanup(Client.java:191)
> at org.apache.hadoop.ipc.Client.stop(Client.java:1235)
> at org.apache.hadoop.ipc.ClientCache.stopClient(ClientCache.java:100)
> at
> org.apache.hadoop.ipc.ProtobufRpcEngine$Invoker.close(ProtobufRpcEngine.java:251)
> at org.apache.hadoop.ipc.RPC.stopProxy(RPC.java:626)
> at
> org.apache.hadoop.yarn.api.impl.pb.client.ApplicationClientProtocolPBClientImpl.close(ApplicationClientProtocolPBClientImpl.java:112)
> at org.apache.hadoop.ipc.RPC.stopProxy(RPC.java:621)
> at
> org.apache.hadoop.io.retry.DefaultFailoverProxyProvider.close(DefaultFailoverProxyProvider.java:57)
> at
> org.apache.hadoop.io.retry.RetryInvocationHandler.close(RetryInvocationHandler.java:206)
> at org.apache.hadoop.ipc.RPC.stopProxy(RPC.java:626)
> at
> org.apache.hadoop.yarn.client.api.impl.YarnClientImpl.serviceStop(YarnClientImpl.java:124)
> at
> org.apache.hadoop.service.AbstractService.stop(AbstractService.java:221)
> . . .
> {code}
> It happens sporadically when stopping YarnClient.
> Looking at the code in Client's 'unrefAndCleanup' its not immediately obvious
> why and who throws the interrupt but in any event it should not be logged as
> ERROR. Probably a WARN with no stack trace.
> Also, for consistency and correctness you may want to Interrupt current
> thread as well.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
