Mailing-List: contact yarn-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: yarn-issues@hadoop.apache.org
Date: Thu, 14 Mar 2013 21:32:16 +0000 (UTC)
From: "Bikas Saha (JIRA)" <jira@apache.org>
To: yarn-issues@hadoop.apache.org
Message-ID: <JIRA.12550672.1334234324274.443289.1363296736038@arcas>
In-Reply-To: <JIRA.12550672.1334234324274@arcas>
References: <JIRA.12550672.1334234324274@arcas>
Subject: [jira] [Commented] (YARN-378) ApplicationMaster retry times should
 be set by Client
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/YARN-378?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13602788#comment-13602788 ] 

Bikas Saha commented on YARN-378:
---------------------------------

I am in favor of setting the value in env in addition to the api. I want it in the api to encourage other app developers to do the desired thing and obtain such (and other) information from the RM upon registration. This is different from the use case of the application attempt id where we need something before contacting the RM.

I also took a quick look at the MR AM code. Its currently reading the value from config and the only use is setting the isLastAMRetry value. The isLastRetry value is later used during job shutdown. Job shutdown will happen after services.start(). So it should not be a terribly large change to get and use the retry value after registration. registration happens during services.start().
                
> ApplicationMaster retry times should be set by Client
> -----------------------------------------------------
>
>                 Key: YARN-378
>                 URL: https://issues.apache.org/jira/browse/YARN-378
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: client, resourcemanager
>         Environment: suse
>            Reporter: xieguiming
>            Assignee: Zhijie Shen
>              Labels: usability
>         Attachments: YARN-378_1.patch, YARN-378_2.patch, YARN-378_3.patch, YARN-378_4.patch, YARN-378_5.patch, YARN-378_6.patch, YARN-378_6.patch, YARN-378_7.patch, YARN-378_8.patch
>
>
> We should support that different client or user have different ApplicationMaster retry times. It also say that "yarn.resourcemanager.am.max-retries" should be set by client. 

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira