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

Posted to issues@tez.apache.org by "Bikas Saha (JIRA)" <ji...@apache.org> on 2014/03/01 03:07:19 UTC

[jira] [Commented] (TEZ-904) Committer recovery events should be out-of-band

    [ https://issues.apache.org/jira/browse/TEZ-904?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13916717#comment-13916717 ] 

Bikas Saha commented on TEZ-904:
--------------------------------

These out of band (or sync) flushes should not stall the dag execution by blocking the async dispatcher. We should be introducing waiting states to make the vertex/dag wait for the flush instead of blocking the threads. 

> Committer recovery events should be out-of-band
> -----------------------------------------------
>
>                 Key: TEZ-904
>                 URL: https://issues.apache.org/jira/browse/TEZ-904
>             Project: Apache Tez
>          Issue Type: Sub-task
>            Reporter: Hitesh Shah
>
> For now, committer events are part of the same event flow as others. Should be handled out of band to ensure a non-flushed event does not cause problems.



--
This message was sent by Atlassian JIRA
(v6.1.5#6160)