[jira] Updated: (HADOOP-1169) CopyFiles skips src files of s3 urls

["stack@archive.org (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HADOOP-1169?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

stack@archive.org updated HADOOP-1169:
--------------------------------------

    Attachment: hadoop1169.patch

> CopyFiles skips src files of s3 urls
> ------------------------------------
>
>                 Key: HADOOP-1169
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1169
>             Project: Hadoop
>          Issue Type: Bug
>          Components: util
>    Affects Versions: 0.12.2
>            Reporter: stack@archive.org
>            Priority: Minor
>         Attachments: hadoop1169.patch
>
>
> When given a source file of items to copy, the CopyFiles utility tries each of its supported schemes in turn looking for matching entries in the passed file.  Of the 4 schemes -- file, hdfs, http, and s3 -- we never make it to the s3 test.  We skip out early.  Also, even if we did make it to the search-for-s3-URLs code, the list of s3 srcPaths would always be empty because of a copy/paste error.

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