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] [Updated] (HDFS-6618) Edit log corruption may still happen even after HDFS-6527
Date Wed, 02 Jul 2014 13:06:25 GMT

     [ https://issues.apache.org/jira/browse/HDFS-6618?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Kihwal Lee updated HDFS-6618:
-----------------------------

    Attachment: HDFS-6618.inodeRemover.v2.patch

Fixed the javadoc warning in v2.

> Edit log corruption may still happen even after HDFS-6527
> ---------------------------------------------------------
>
>                 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
>            Priority: Blocker
>         Attachments: HDFS-6618.AbstractList.patch, HDFS-6618.inodeRemover.patch, HDFS-6618.inodeRemover.v2.patch,
HDFS-6618.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