hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Pete Wyckoff (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HADOOP-4398) fuse-dfs per FD context is not thread safe and can cause segfaults and corruptions
Date Fri, 17 Oct 2008 00:16:44 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-4398?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12640384#action_12640384
] 

Pete Wyckoff commented on HADOOP-4398:
--------------------------------------

core test failure not due to this patch as fuse-dfs is not accessed in any way by any of those
tests

> fuse-dfs per FD context is not thread safe and can cause segfaults and corruptions
> ----------------------------------------------------------------------------------
>
>                 Key: HADOOP-4398
>                 URL: https://issues.apache.org/jira/browse/HADOOP-4398
>             Project: Hadoop Core
>          Issue Type: Bug
>          Components: contrib/fuse-dfs
>    Affects Versions: 0.20.0
>            Reporter: Pete Wyckoff
>            Assignee: Pete Wyckoff
>             Fix For: 0.20.0
>
>         Attachments: HADOOP-4398.1.txt, HADOOP-4398.2.txt
>
>
> this affects dfs_read and dfs_write and dfs_open dfs_release.
> people have seen corruptions in 0.18.1 on reads - hadoop-4397.
> In 0.19 since there is no global DFS handle, it is slightly different, but the same principles
need to be applied and make the per FD context data (ie ptr to DFS, read buffer, ...) thread
safe.
> I can't think of a way to add a reliable test case for this in the current framework,
but should be sure to assert pre/post conditions wherever possible.

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


Mime
View raw message