hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Prabhu Joseph (JIRA)" <j...@apache.org>
Subject [jira] [Created] (YARN-5933) ATS stale entries in active directory causes ApplicationNotFoundException in RM
Date Thu, 24 Nov 2016 04:27:58 GMT
Prabhu Joseph created YARN-5933:
-----------------------------------

             Summary: ATS stale entries in active directory causes ApplicationNotFoundException
in RM
                 Key: YARN-5933
                 URL: https://issues.apache.org/jira/browse/YARN-5933
             Project: Hadoop YARN
          Issue Type: Bug
          Components: ATSv2
    Affects Versions: 2.7.3
            Reporter: Prabhu Joseph


On Secure cluster where ATS is down, Tez job submitted will fail while getting TIMELINE_DELEGATION_TOKEN
with below exception

{code}
0: jdbc:hive2://kerberos-2.openstacklocal:100> select csmallint from alltypesorc group
by csmallint;
INFO  : Session is already open
INFO  : Dag name: select csmallint from alltypesor...csmallint(Stage-1)
INFO  : Tez session was closed. Reopening...
ERROR : Failed to execute tez graph.
java.lang.RuntimeException: Failed to connect to timeline server. Connection retries limit
exceeded. The posted timeline event may be missing
	at org.apache.hadoop.yarn.client.api.impl.TimelineClientImpl$TimelineClientConnectionRetry.retryOn(TimelineClientImpl.java:266)
	at org.apache.hadoop.yarn.client.api.impl.TimelineClientImpl.operateDelegationToken(TimelineClientImpl.java:590)
	at org.apache.hadoop.yarn.client.api.impl.TimelineClientImpl.getDelegationToken(TimelineClientImpl.java:506)
	at org.apache.hadoop.yarn.client.api.impl.YarnClientImpl.getTimelineDelegationToken(YarnClientImpl.java:349)
	at org.apache.hadoop.yarn.client.api.impl.YarnClientImpl.addTimelineDelegationToken(YarnClientImpl.java:330)
	at org.apache.hadoop.yarn.client.api.impl.YarnClientImpl.submitApplication(YarnClientImpl.java:250)
	at org.apache.tez.client.TezYarnClient.submitApplication(TezYarnClient.java:72)
	at org.apache.tez.client.TezClient.start(TezClient.java:409)
	at org.apache.hadoop.hive.ql.exec.tez.TezSessionState.open(TezSessionState.java:196)
	at org.apache.hadoop.hive.ql.exec.tez.TezSessionPoolManager.closeAndOpen(TezSessionPoolManager.java:311)
	at org.apache.hadoop.hive.ql.exec.tez.TezTask.submit(TezTask.java:453)
	at org.apache.hadoop.hive.ql.exec.tez.TezTask.execute(TezTask.java:180)
	at org.apache.hadoop.hive.ql.exec.Task.executeTask(Task.java:160)
	at org.apache.hadoop.hive.ql.exec.TaskRunner.runSequential(TaskRunner.java:89)
	at org.apache.hadoop.hive.ql.Driver.launchTask(Driver.java:1728)
	at org.apache.hadoop.hive.ql.Driver.execute(Driver.java:1485)
	at org.apache.hadoop.hive.ql.Driver.runInternal(Driver.java:1262)
	at org.apache.hadoop.hive.ql.Driver.run(Driver.java:1126)
	at org.apache.hadoop.hive.ql.Driver.run(Driver.java:1121)
	at org.apache.hive.service.cli.operation.SQLOperation.runQuery(SQLOperation.java:154)
	at org.apache.hive.service.cli.operation.SQLOperation.access$100(SQLOperation.java:71)
	at org.apache.hive.service.cli.operation.SQLOperation$1$1.run(SQLOperation.java:206)
	at java.security.AccessController.doPrivileged(Native Method)
	at javax.security.auth.Subject.doAs(Subject.java:422)
	at org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1709)
	at org.apache.hive.service.cli.operation.SQLOperation$1.run(SQLOperation.java:218)
	at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
	at java.util.concurrent.FutureTask.run(FutureTask.java:266)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
	at java.lang.Thread.run(Thread.java:745)
{code}

Tez YarnClient has received an applicationID from RM. On Restarting ATS now, ATS tries to
get the application report from RM and so RM will throw ApplicationNotFoundException. ATS
will keep on requesting and which floods RM.

{code}
RM logs:
2016-11-23 13:53:57,345 INFO org.apache.hadoop.yarn.server.resourcemanager.ClientRMService:
Allocated new applicationId: 5
2016-11-23 14:05:04,936 INFO org.apache.hadoop.ipc.Server: IPC Server handler 9 on 8050, call
org.apache.hadoop.yarn.api.ApplicationClientProtocolPB.getApplicationReport from 172.26.71.120:37699
Call#26 Retry#0
org.apache.hadoop.yarn.exceptions.ApplicationNotFoundException: Application with id 'application_1479897867169_0005'
doesn't exist in RM.
	at org.apache.hadoop.yarn.server.resourcemanager.ClientRMService.getApplicationReport(ClientRMService.java:328)
	at org.apache.hadoop.yarn.api.impl.pb.service.ApplicationClientProtocolPBServiceImpl.getApplicationReport(ApplicationClientProtocolPBServiceImpl.java:175)
	at org.apache.hadoop.yarn.proto.ApplicationClientProtocol$ApplicationClientProtocolService$2.callBlockingMethod(ApplicationClientProtocol.java:417)
	at org.apache.hadoop.ipc.ProtobufRpcEngine$Server$ProtoBufRpcInvoker.call(ProtobufRpcEngine.java:616)
	at org.apache.hadoop.ipc.RPC$Server.call(RPC.java:969)
	at org.apache.hadoop.ipc.Server$Handler$1.run(Server.java:2206)
	at org.apache.hadoop.ipc.Server$Handler$1.run(Server.java:2202)
	at java.security.AccessController.doPrivileged(Native Method)
	at javax.security.auth.Subject.doAs(Subject.java:422)
	at org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1709)
	at org.apache.hadoop.ipc.Server$Handler.run(Server.java:2200)
{code}

There is a stale application entry inside /ats/active directory. ATS stops requesting when
we remove this directory.

[hive@kerberos-2 bin]$ hadoop fs -ls /ats/active
drwxrwx---   - hive hadoop          0 2016-11-23 13:54 /ats/active/application_1479897867169_0005

This issue with ATS is exposed by Tez job as Tez uses putDomain method. On calling TimelineClientImpl#putDomain()
-> writeDomain() -> getAppAttemptDir() -> createApplicationDir() which creates a
application directory inside ATS activePath. After Tez job created this, it fails as unable
to connect to ATS. Now when ATS comes back, it scans activePath for every 60 seconds (yarn.timeline-service.entity-group-fs-store.scan-interval-seconds)
and calls GetApplicationReport which leads to ApplicationNotFoundException in RM. 

For this negative case - we can delete the appDirectory inside activePath from ATS EntityGroupFSTimelineStore#getAppState()
once the RM throws ApplicationNotFoundException.






--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: yarn-issues-help@hadoop.apache.org


Mime
View raw message