hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Eli Collins (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HDFS-2186) DN volume failures on startup are not counted
Date Fri, 22 Jul 2011 07:28:57 GMT

     [ https://issues.apache.org/jira/browse/HDFS-2186?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel

Eli Collins updated HDFS-2186:

    Attachment: hdfs-2186-1.patch

Patch attached. Considers any configured volume that is not a valid storage directory to be
a failed volume. The new test asserts that a failed volume on startup is seen as such by the
NN. This is visible in the web UI "failed volumes" field now, ie a DN that starts with 1 failed
volume will show 1 failed volume on the web UI.

> DN volume failures on startup are not counted
> ---------------------------------------------
>                 Key: HDFS-2186
>                 URL: https://issues.apache.org/jira/browse/HDFS-2186
>             Project: Hadoop HDFS
>          Issue Type: Bug
>          Components: data-node
>            Reporter: Eli Collins
>             Fix For: 0.23.0
>         Attachments: hdfs-2186-1.patch
> Volume failures detected on startup are not currently counted/reported as such. Eg if
you have configured 4 volumes, 2 tolerated failures, and you start a DN with two failed volumes
it will come up and report (to the NN) no failed volumes. The DN will still be able to tolerate
2 additional volume failures (ie it's OK with no valid volumes remaining). The intent of the
volume failure toleration config value is that if more than this # of volumes of the total
set of configured volumes have failed the DN should shutdown, therefore volume failures detected
on startup should count against this quota. 

This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira


View raw message