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

Posted to dev@hive.apache.org by "Yash Datta (JIRA)" <ji...@apache.org> on 2014/04/08 06:19:14 UTC

[jira] [Updated] (HIVE-6490) Hooks to map reduce task output (OutputCommitter) for Hive Storage Handlers

     [ https://issues.apache.org/jira/browse/HIVE-6490?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Yash Datta updated HIVE-6490:
-----------------------------

    Description: 
If user sets the "mapred.output.committer.class" property in jobConf via the storage handler implementation , then he/she can override the nulloutputcommitter to get hooks for commitJob , cleanupJob etc.

Hence we can make use of the outputcommitter class in hive :
http://hadoop.apache.org/docs/r2.2.0/api/org/apache/hadoop/mapred/OutputCommitter.html

commitJob provides a place to do job context tasks if needed by the storage handler 

  was:
If user sets the "mapred.output.committer.class" property in jobConf via the storage handler implementation , then he/she can override the nulloutputcommitter to get hooks for commitJob , cleanupJob etc.

commitJob provides a place to do job context tasks if needed by the storage handler 

        Summary: Hooks to map reduce task output (OutputCommitter) for Hive Storage Handlers  (was: Override the default NullOutputCommitter in hive hadoop shims class)

> Hooks to map reduce task output (OutputCommitter) for Hive Storage Handlers
> ---------------------------------------------------------------------------
>
>                 Key: HIVE-6490
>                 URL: https://issues.apache.org/jira/browse/HIVE-6490
>             Project: Hive
>          Issue Type: Improvement
>          Components: Configuration
>    Affects Versions: 0.10.0
>            Reporter: Yash Datta
>            Priority: Minor
>              Labels: patch
>             Fix For: 0.10.1
>
>         Attachments: HIVE-6490-branch-0.10.patch
>
>
> If user sets the "mapred.output.committer.class" property in jobConf via the storage handler implementation , then he/she can override the nulloutputcommitter to get hooks for commitJob , cleanupJob etc.
> Hence we can make use of the outputcommitter class in hive :
> http://hadoop.apache.org/docs/r2.2.0/api/org/apache/hadoop/mapred/OutputCommitter.html
> commitJob provides a place to do job context tasks if needed by the storage handler 



--
This message was sent by Atlassian JIRA
(v6.2#6252)