sqoop-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jarek Jarcec Cecho (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (SQOOP-1631) Staging-table name check don't need using PGBulkloadManager
Date Sun, 07 Dec 2014 16:26:12 GMT

    [ https://issues.apache.org/jira/browse/SQOOP-1631?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14237196#comment-14237196
] 

Jarek Jarcec Cecho commented on SQOOP-1631:
-------------------------------------------

It seems that the only usage of the {{--clear-staging-table}} parameter inside the PGBulkload
connector is to [drop the staging table if it exists|https://github.com/apache/sqoop/blob/trunk/src/java/org/apache/sqoop/mapreduce/postgresql/PGBulkloadExportMapper.java#L86].
I do feel that this is quite overloading the semantics of the parameter and hence I would
much rather see if the connector would stop using it this way. User can always set the {{pgbulkload.clear.staging.table}}
on the command line while executing Sqoop.

> Staging-table name check don't need using PGBulkloadManager
> -----------------------------------------------------------
>
>                 Key: SQOOP-1631
>                 URL: https://issues.apache.org/jira/browse/SQOOP-1631
>             Project: Sqoop
>          Issue Type: Bug
>          Components: connectors/postgresql
>    Affects Versions: 1.4.6
>            Reporter: Masahiro Yamaguchi
>         Attachments: SQOOP-1631_1.patch, SQOOP-1631_2.patch
>
>
> If we use PGBulkloadManager, the number of staging-tables  equals number of map tasks.
> And staging-table's name is decided as follows.
>   tmpTableName = table + "_" + context.getTaskAttemptID().toString();
> ExportTools check staging-table's name, but this check don't need using PGBulkloadManager.



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

Mime
View raw message