Mailing-List: contact yarn-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Date: Tue, 10 May 2016 05:54:12 +0000 (UTC)
From: "Rohith Sharma K S (JIRA)" <jira@apache.org>
To: yarn-issues@hadoop.apache.org
Message-ID: <JIRA.12966052.1462798505000.150506.1462859652942@Atlassian.JIRA>
In-Reply-To: <JIRA.12966052.1462798505000@Atlassian.JIRA>
References: <JIRA.12966052.1462798505000@Atlassian.JIRA> <JIRA.12966052.1462798505473@arcas>
Subject: [jira] [Commented] (YARN-5063) Fail to launch AM continuously on a
 lost NM
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Tue, 10 May 2016 05:54:14 -0000


    [ https://issues.apache.org/jira/browse/YARN-5063?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15277664#comment-15277664 ] 

Rohith Sharma K S commented on YARN-5063:
-----------------------------------------

Thanks for clarifying my doubts. I think this is very good scenario for AM node blacklisting to be consider. Keeping cc:/ [~sunilg] [~vvasudev] [~vinodkv] let also they know about scenario. As I said, there are design level issue in YARN-2005, so need to wait for proper solution. Ex : Node is not reachable so that node is blacklisted. Since other nodes are busy, the same  node comes back(registered) where in new node is not considered for allocation. See YARN-4685

> Fail to launch AM continuously on a lost NM
> -------------------------------------------
>
>                 Key: YARN-5063
>                 URL: https://issues.apache.org/jira/browse/YARN-5063
>             Project: Hadoop YARN
>          Issue Type: Bug
>          Components: resourcemanager
>            Reporter: Jun Gong
>            Assignee: Jun Gong
>
> If a NM node shuts down, RM will not mark it as LOST until liveness monitor finds it timeout. However before that, RM might continuously allocate AM on that NM.
> We found this case in our cluster: RM continuously allocated a same AM on a lost NM before RM found it lost, and AMLauncher always failed because it could not connect to the lost NM. To solve the problem, we could add the NM to AM blacklist if RM failed to launch it.


--
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