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

Posted to mapreduce-issues@hadoop.apache.org by "Gelesh (JIRA)" <ji...@apache.org> on 2013/04/08 06:25:16 UTC

[jira] [Commented] (MAPREDUCE-4879) TeraOutputFormat may overwrite an existing output directory

    [ https://issues.apache.org/jira/browse/MAPREDUCE-4879?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13625123#comment-13625123 ] 

Gelesh commented on MAPREDUCE-4879:
-----------------------------------

[~jira.shegalov]
Since TeraOutputFormat extends FileOutputFormat, it must be in compliance with checkOutputSpecs(), of FileOutputFormat. And checkOutputSpecs() is supposed to be called from Job Client. If there is any issue, I feel it would be better to fix it within the existing control flow.
                
> TeraOutputFormat may overwrite an existing output directory
> -----------------------------------------------------------
>
>                 Key: MAPREDUCE-4879
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-4879
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: examples
>    Affects Versions: trunk
>            Reporter: Gera Shegalov
>         Attachments: MAPREDUCE-4879-trunk.patch, MAPREDUCE-4879-trunk-rev1.patch
>
>
> Unlike FileOutputFormat, TeraOutputFormat does not prevent TeraGen/Sort jobs from writing into an existing directory, and potentially overwriting previous runs.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira