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

Posted to issues@tez.apache.org by "Rajesh Balamohan (JIRA)" <ji...@apache.org> on 2015/06/30 06:44:04 UTC

[jira] [Created] (TEZ-2584) Investigate: Reducers with local errors can run indefinitely

Rajesh Balamohan created TEZ-2584:
-------------------------------------

             Summary: Investigate: Reducers with local errors can run indefinitely
                 Key: TEZ-2584
                 URL: https://issues.apache.org/jira/browse/TEZ-2584
             Project: Apache Tez
          Issue Type: Bug
            Reporter: Rajesh Balamohan


When local errors (e.g disk error) are there, fetchers are expected to throw error and it is expected to restart the task elsewhere.  With master branch, reducer continues to run indefinitely.  Creating this jira to investigate the issue. I don't have a test case for this, but can be easily be simulated by throwing IOException purposefully in FetcherOrderedGrouped-->copyMapOutput (reserve()). 



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)