hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sunil G (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-5333) Some recovered apps are put into default queue when RM HA
Date Thu, 04 Aug 2016 05:47:20 GMT

    [ https://issues.apache.org/jira/browse/YARN-5333?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15407212#comment-15407212
] 

Sunil G commented on YARN-5333:
-------------------------------

Hi [~jianhe] and [~rohithsharma] [~hex108]

bq.I think the ActiveStandbyElector will handle reJoin automatically if exception thrown from
the transtionToActive method.We can use the fail fast config for this scenario.
I have another view here. If {{refreshAll}} fails due to a corrupted capacity-scheduler.xml
file when a transitionToActive is happening (IOException will come from *refreshQeues*) as
mentioned in YARN-3893, I think we need to fail-fast. Yes, it will be harsh. But we can avoid
a scenario like both RMs in standby (and both RMs will switchover continuously) if I am not
wrong. pls correct me if I understood this case wrong. This will happen if config is given
not-to fail fast. pls share your thoughts.

> 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, YARN-5333.02.patch, YARN-5333.03.patch, YARN-5333.04.patch,
YARN-5333.05.patch, YARN-5333.06.patch, YARN-5333.07.patch, YARN-5333.08.patch, YARN-5333.09.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


Mime
View raw message