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

Posted to dev@pig.apache.org by "Olga Natkovich (JIRA)" <ji...@apache.org> on 2011/02/24 02:25:38 UTC

[jira] Updated: (PIG-1556) Need a clean way to kill Pig jobs.

     [ https://issues.apache.org/jira/browse/PIG-1556?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Olga Natkovich updated PIG-1556:
--------------------------------

    Fix Version/s:     (was: 0.9.0)

> Need a clean way to kill Pig jobs.
> ----------------------------------
>
>                 Key: PIG-1556
>                 URL: https://issues.apache.org/jira/browse/PIG-1556
>             Project: Pig
>          Issue Type: New Feature
>          Components: tools
>    Affects Versions: 0.7.0
>            Reporter: Aravind Srinivasan
>
> We need a way to kill a running Pig script cleanly. This is very similar to " hadoop job -kill" command. This requirement means the following. 
> 1) Support a "pig -kill <script ID>" or a similar syntax. The script ID or some unique handle should be easily available for the user to identify a running Pig job.
> 2) The command will then identify all the MR jobs that are currently spawned by this given Pig script.
> 3) It will internally usse "hadoop job -kill" to kill each one of those MR jobs spawned.
> 4) It will do any other cleanup necessary and also make sure all mappers/reducers emanating from this Pig script are killed.

-- 
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira