[
https://issues.apache.org/jira/browse/HADOOP-13738?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15606082#comment-15606082
]
Arpit Agarwal commented on HADOOP-13738:
----------------------------------------
[~kihwal] that's a good point. The caller should deal with slow checks since a
single check can also hang indefinitely. The DN currently serializes disk
checks so one slow disk can delay checking the rest. Also it doesn't handle
stalled checks. I am testing some changes to fix this in the DataNode and plan
to post patches in the next few days.
> DiskChecker should perform some disk IO
> ---------------------------------------
>
> Key: HADOOP-13738
> URL: https://issues.apache.org/jira/browse/HADOOP-13738
> Project: Hadoop Common
> Issue Type: Improvement
> Reporter: Arpit Agarwal
> Assignee: Arpit Agarwal
> Attachments: HADOOP-13738.01.patch, HADOOP-13738.02.patch,
> HADOOP-13738.03.patch
>
>
> DiskChecker can fail to detect total disk/controller failures indefinitely.
> We have seen this in real clusters. DiskChecker performs simple
> permissions-based checks on directories which do not guarantee that any disk
> IO will be attempted.
> A simple improvement is to write some data and flush it to the disk.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
