[
https://issues.apache.org/jira/browse/SQOOP-3136?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15875080#comment-15875080
]
Illya Yalovyy commented on SQOOP-3136:
--------------------------------------
[~jarcec], could you please clarify acceptance process for a patch? Is anything
expecting from me? Who would you recommend to add to CR to get it going?
Thanks,
Illya
> Sqoop should work well with not default file systems
> ----------------------------------------------------
>
> Key: SQOOP-3136
> URL: https://issues.apache.org/jira/browse/SQOOP-3136
> Project: Sqoop
> Issue Type: Improvement
> Components: connectors/hdfs
> Affects Versions: 1.4.5
> Reporter: Illya Yalovyy
> Assignee: Illya Yalovyy
> Attachments: SQOOP-3136.patch
>
>
> Currently Sqoop assumes default file system when it comes to IO operations.
> It makes it hard to use other FileSystem implementations as source or
> destination. Here is an example:
> {code}
> sqoop import --connect <JDBC CONNECTION> --table table1 --driver <JDBC
> DRIVER> --username root --password **** --delete-target-dir --target-dir
> s3a://some-bucket/tmp/sqoop
> ...
> 17/02/15 19:16:59 ERROR tool.ImportTool: Imported Failed: Wrong FS:
> s3a://some-bucket/tmp/sqoop, expected: hdfs://<DNS>:8020
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
