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 "Joe Crobak (JIRA)" <ji...@apache.org> on 2009/12/12 20:51:18 UTC

[jira] Updated: (MAPREDUCE-1290) DBOutputFormat does not support rewriteBatchedStatements when using MySQL jdbc drivers

     [ https://issues.apache.org/jira/browse/MAPREDUCE-1290?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Joe Crobak updated MAPREDUCE-1290:
----------------------------------

    Attachment: MAPREDUCE-1290.patch

I'm not aware of an appropriate way to test the submitted patch, because the hadoop tests using HSQL don't seem to mind the stray semicolon. 

> DBOutputFormat does not support rewriteBatchedStatements when using MySQL jdbc drivers
> --------------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-1290
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-1290
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>            Reporter: Joe Crobak
>         Attachments: MAPREDUCE-1290.patch
>
>
> The DBOutputFormat adds a semi-colon to the end of the INSERT statement that it uses to save fields to the database.  Semicolons are typically used in command line programs but are not needed when using the JDBC API.  In this case, the stray semi-colon breaks rewriteBatchedStatement support. See: http://forums.mysql.com/read.php?39,271526,271526#msg-271526 for an example.
> In my use case, rewriteBatchedStatement is very useful because it increases the speed of inserts and reduces memory consumption.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.