Mailing-List: contact yarn-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Date: Tue, 12 Jul 2016 16:02:20 +0000 (UTC)
From: "Jun Gong (JIRA)" <jira@apache.org>
To: yarn-issues@hadoop.apache.org
Message-ID: <JIRA.12987470.1467895856000.4041.1468339340595@Atlassian.JIRA>
In-Reply-To: <JIRA.12987470.1467895856000@Atlassian.JIRA>
References: <JIRA.12987470.1467895856000@Atlassian.JIRA> <JIRA.12987470.1467895856951@arcas>
Subject: [jira] [Updated] (YARN-5333) Some recovered apps are put into
 default queue when RM HA
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Tue, 12 Jul 2016 16:02:23 -0000


     [ https://issues.apache.org/jira/browse/YARN-5333?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Jun Gong updated YARN-5333:
---------------------------
    Description: 
Enable RM HA and use FairScheduler, {{yarn.scheduler.fair.allow-undeclared-pools}} is set to false, {{yarn.scheduler.fair.user-as-default-queue}} is set to false.

Reproduce steps:
1. Start two RMs.
2. After RMs are running, change both RM's file {{etc/hadoop/fair-scheduler.xml}}, then add some queues.
3. Submit some apps to the new added queues.
4. Stop the active RM, then the standby RM will transit to active and recover apps.
However the new active RM will put recovered apps into default queue because it might have not loaded the new {{fair-scheduler.xml}}. We need call {{initScheduler}} before start active services or bring {{refreshAll()}} in front of {{rm.transitionToActive()}}. *It seems it is also important for other scheduler*.

  was:
Enable RM HA and use FairScheduler, {{yarn.scheduler.fair.allow-undeclared-pools}} is set to false, {{yarn.scheduler.fair.user-as-default-queue}} is set to false.

Reproduce steps:
1. Start two RMs.
2. After RMs are running, change both RM's file {{etc/hadoop/fair-scheduler.xml}}, then add some queues.
3. Submit some apps to the new added queues.
4. Stop the active RM, then the standby RM will transit to active and recover apps.
However the new active RM will reject recovered apps because it might have not loaded the new {{fair-scheduler.xml}}. We need call {{initScheduler}} before start active services or bring {{refreshAll()}} in front of {{rm.transitionToActive()}}. *It seems it is aslo important for other scheduler*.

Related logs are as following:
{quote}
2016-07-07 16:55:34,756 INFO org.apache.hadoop.yarn.server.resourcemanager.ResourceManager: Recover ended
...
2016-07-07 16:55:34,824 INFO org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.AllocationFileLoaderService: Loading allocation file /gaia/hadoop/etc/hadoop/fair-scheduler.xml
2016-07-07 16:55:34,826 ERROR org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.FairScheduler: Application rejected by queue placement policy
2016-07-07 16:55:34,828 INFO org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.FairScheduler: Application appattempt_1467803586002_0006_000001 is done. finalState=FAILED
2016-07-07 16:55:34,828 INFO org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.FairScheduler: Unknown application appattempt_1467803586002_0006_000001 has completed!
2016-07-07 16:55:34,828 ERROR org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.FairScheduler: Application rejected by queue placement policy
2016-07-07 16:55:34,828 INFO org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.FairScheduler: Application appattempt_1467803586002_0004_000001 is done. finalState=FAILED
2016-07-07 16:55:34,828 INFO org.apache.hadoop.yarn.server.resourcemanager.scheduler.fair.FairScheduler: Unknown application appattempt_1467803586002_0004_000001 has completed!
2016-07-07 16:55:34,828 ERROR org.apache.hadoop.yarn.server.resourcemanager.rmapp.RMAppImpl: Can't handle this event at current state
org.apache.hadoop.yarn.state.InvalidStateTransitonException: Invalid event: APP_REJECTED at ACCEPTED
	at org.apache.hadoop.yarn.state.StateMachineFactory.doTransition(StateMachineFactory.java:305)
	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.RMAppImpl.handle(RMAppImpl.java:697)
	at org.apache.hadoop.yarn.server.resourcemanager.rmapp.RMAppImpl.handle(RMAppImpl.java:88)
	at org.apache.hadoop.yarn.server.resourcemanager.ResourceManager$ApplicationEventDispatcher.handle(ResourceManager.java:718)
	at org.apache.hadoop.yarn.server.resourcemanager.ResourceManager$ApplicationEventDispatcher.handle(ResourceManager.java:702)
	at org.apache.hadoop.yarn.event.AsyncDispatcher.dispatch(AsyncDispatcher.java:191)
	at org.apache.hadoop.yarn.event.AsyncDispatcher$1.run(AsyncDispatcher.java:124)
	at java.lang.Thread.run(Thread.java:745)
{quote}


> Some recovered apps are put into default queue when RM HA
> ---------------------------------------------------------
>
>                 Key: YARN-5333
>                 URL: https://issues.apache.org/jira/browse/YARN-5333
>             Project: Hadoop YARN
>          Issue Type: Bug
>            Reporter: Jun Gong
>            Assignee: Jun Gong
>         Attachments: YARN-5333.01.patch
>
>
> Enable RM HA and use FairScheduler, {{yarn.scheduler.fair.allow-undeclared-pools}} is set to false, {{yarn.scheduler.fair.user-as-default-queue}} is set to false.
> Reproduce steps:
> 1. Start two RMs.
> 2. After RMs are running, change both RM's file {{etc/hadoop/fair-scheduler.xml}}, then add some queues.
> 3. Submit some apps to the new added queues.
> 4. Stop the active RM, then the standby RM will transit to active and recover apps.
> However the new active RM will put recovered apps into default queue because it might have not loaded the new {{fair-scheduler.xml}}. We need call {{initScheduler}} before start active services or bring {{refreshAll()}} in front of {{rm.transitionToActive()}}. *It seems it is also important for other scheduler*.


--
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