Mailing-List: contact yarn-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Date: Fri, 21 Oct 2016 15:58:58 +0000 (UTC)
From: "Rohith Sharma K S (JIRA)" <jira@apache.org>
To: yarn-issues@hadoop.apache.org
Message-ID: <JIRA.13014002.1476987728000.52965.1477065538957@Atlassian.JIRA>
In-Reply-To: <JIRA.13014002.1476987728000@Atlassian.JIRA>
References: <JIRA.13014002.1476987728000@Atlassian.JIRA> <JIRA.13014002.1476987728398@arcas>
Subject: [jira] [Commented] (YARN-5759) Capability to register for a
 notification/callback on the expiry of timeouts
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Fri, 21 Oct 2016 15:59:01 -0000


    [ https://issues.apache.org/jira/browse/YARN-5759?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15595494#comment-15595494 ] 

Rohith Sharma K S commented on YARN-5759:
-----------------------------------------

I believe intention of notification/callback is to clean up AM specific data which are stored in any of the storage. Unfortunately, there is no such hook or alternative solution exist today. There are couple of JIRA exist to provide solution in-general. YARN-2261 is Umbrella JIRA trying to provide  a way to run post-application cleanup. YARN-5658 is also another related JIRA. 

> Capability to register for a notification/callback on the expiry of timeouts
> ----------------------------------------------------------------------------
>
>                 Key: YARN-5759
>                 URL: https://issues.apache.org/jira/browse/YARN-5759
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: resourcemanager
>            Reporter: Gour Saha
>
> There is a need for the YARN native services REST-API service, to take certain actions once a timeout of an application expires. For example, an immediate requirement is to destroy a Slider application, once its lifetime timeout expires and YARN has stopped the application. Destroying a Slider application means cleanup of Slider HDFS state store and ZK paths for that application. 
> Potentially, there will be advanced requirements from the REST-API service and other services in the future, which will make this feature very handy.


--
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