hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Harsh J (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-11422) Need a method to refresh the list of NN StorageDirectories after removal
Date Fri, 17 Feb 2017 03:24:41 GMT

    [ https://issues.apache.org/jira/browse/HDFS-11422?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15871115#comment-15871115

Harsh J commented on HDFS-11422:

Does the feature offered by {{dfs.namenode.name.dir.restore}} help, even partially? For NNs
with multiple directories, a reattempt of an earlier failed directory gets tried at every
checkpoint trigger. Its not purely time based however.

We'd considered making it default to true (HDFS-3560), but that was not done. If this fits,
perhaps we can turn it on by default as the feature has existed for a while now.

> Need a method to refresh the list of NN StorageDirectories after removal
> ------------------------------------------------------------------------
>                 Key: HDFS-11422
>                 URL: https://issues.apache.org/jira/browse/HDFS-11422
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: namenode
>    Affects Versions: 2.6.0
>            Reporter: Attila Bukor
> If a NN storage directory is removed due to an error, the NameNode will fail to write
the image even if the issue was intermittent. It would be good to have a way to make the NameNode
try writing again after the issue is fixed - and maybe even try it automatically every certain
amount of time (configurable).

This message was sent by Atlassian JIRA

To unsubscribe, e-mail: hdfs-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-help@hadoop.apache.org

View raw message