hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Kihwal Lee (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-6618) Remove deleted INodes from INodeMap right away
Date Wed, 09 Jul 2014 17:51:06 GMT

    [ https://issues.apache.org/jira/browse/HDFS-6618?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14056512#comment-14056512
] 

Kihwal Lee commented on HDFS-6618:
----------------------------------

The build failed because of this. 
{panel}
     [exec] CMake Error at /usr/share/cmake-2.8/Modules/FindPackageHandleStandardArgs.cmake:108
(message):
     [exec]   Could NOT find PkgConfig (missing: PKG_CONFIG_EXECUTABLE)
     [exec] Call Stack (most recent call first):
     [exec]   /usr/share/cmake-2.8/Modules/FindPackageHandleStandardArgs.cmake:315 (_FPHSA_FAILURE_MESSAGE)
     [exec]   /usr/share/cmake-2.8/Modules/FindPkgConfig.cmake:106 (find_package_handle_standard_args)
     [exec]   main/native/fuse-dfs/CMakeLists.txt:23 (find_package)
{panel}



> Remove deleted INodes from INodeMap right away
> ----------------------------------------------
>
>                 Key: HDFS-6618
>                 URL: https://issues.apache.org/jira/browse/HDFS-6618
>             Project: Hadoop HDFS
>          Issue Type: Bug
>    Affects Versions: 2.5.0
>            Reporter: Kihwal Lee
>            Assignee: Kihwal Lee
>            Priority: Blocker
>         Attachments: HDFS-6618.AbstractList.patch, HDFS-6618.inodeRemover.patch, HDFS-6618.inodeRemover.v2.patch,
HDFS-6618.patch, HDFS-6618.simpler.patch
>
>
> After HDFS-6527, we have not seen the edit log corruption for weeks on multiple clusters
until yesterday. Previously, we would see it within 30 minutes on a cluster.
> But the same condition was reproduced even with HDFS-6527.  The only explanation is that
the RPC handler thread serving {{addBlock()}} was accessing stale parent value.  Although
nulling out parent is done inside the {{FSNamesystem}} and {{FSDirectory}} write lock, there
is no memory barrier because there is no "synchronized" block involved in the process.
> I suggest making parent volatile.



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Mime
View raw message