You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@bigtop.apache.org by "RJ Nowling (JIRA)" <ji...@apache.org> on 2015/02/17 16:17:12 UTC

[jira] [Assigned] (BIGTOP-1544) [BigPetStore] Use of java.io.File and check for directory existence will fail with non-POSIX DFSs (HDFS, S3, etc.)

     [ https://issues.apache.org/jira/browse/BIGTOP-1544?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

RJ Nowling reassigned BIGTOP-1544:
----------------------------------

    Assignee: RJ Nowling

> [BigPetStore] Use of java.io.File and check for directory existence will fail with non-POSIX DFSs (HDFS, S3, etc.)
> ------------------------------------------------------------------------------------------------------------------
>
>                 Key: BIGTOP-1544
>                 URL: https://issues.apache.org/jira/browse/BIGTOP-1544
>             Project: Bigtop
>          Issue Type: Bug
>          Components: blueprints
>            Reporter: RJ Nowling
>            Assignee: RJ Nowling
>             Fix For: 0.9.0
>
>         Attachments: bps_spark_rm_dir_exist_check.patch
>
>
> The BigPetStore Spark driver uses a java.io.File object to store the output directory path. The File's methods are used to check if the path exists and is a directory.
> Since non-POSIX DFSs (S3, HDFS, etc.) will not have paths of the appropriate format and will not be accessible through a POSIX interface, these checks will fail, preventing the usage of S3, HDFS, etc.



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