You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@flink.apache.org by "zhijiang (JIRA)" <ji...@apache.org> on 2017/07/10 07:43:01 UTC

[jira] [Commented] (FLINK-6161) Retry connection in case of a ResourceManager heartbeat timeout

    [ https://issues.apache.org/jira/browse/FLINK-6161?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16079979#comment-16079979 ] 

zhijiang commented on FLINK-6161:
---------------------------------

[~till.rohrmann], would you mind assigning this issue to me?  I am clear of the requirements from above description, so I think I can handle it this week if no other concerns with this issue.:)


> Retry connection in case of a ResourceManager heartbeat timeout
> ---------------------------------------------------------------
>
>                 Key: FLINK-6161
>                 URL: https://issues.apache.org/jira/browse/FLINK-6161
>             Project: Flink
>          Issue Type: Sub-task
>          Components: Distributed Coordination
>    Affects Versions: 1.3.0
>            Reporter: Till Rohrmann
>              Labels: flip-6
>             Fix For: 1.4.0
>
>
> The {{JobMaster}} should try reconnecting to the latest known resource manager leader in case of a resource manager heartbeat timeout. Otherwise the {{JobMaster}} will only try connecting to a {{ResourceManager}} if the leader address information change. In case of a false positive heartbeat timeout, this could break the {{JobMaster's}} connection to the {{ResourceManager}} permanently.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)