hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Andrew Wang (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-7409) Allow dead nodes to finish decommissioning if all files are fully replicated
Date Wed, 19 Nov 2014 00:01:33 GMT

    [ https://issues.apache.org/jira/browse/HDFS-7409?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14217082#comment-14217082
] 

Andrew Wang commented on HDFS-7409:
-----------------------------------

[~mingma] this is basically the approach you recommended with looking at the pending and under-replicated
count. I'm also looking into maintaining that block list too, as it's a stronger solution.

> Allow dead nodes to finish decommissioning if all files are fully replicated
> ----------------------------------------------------------------------------
>
>                 Key: HDFS-7409
>                 URL: https://issues.apache.org/jira/browse/HDFS-7409
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>    Affects Versions: 2.5.1
>            Reporter: Andrew Wang
>            Assignee: Andrew Wang
>            Priority: Minor
>         Attachments: hdfs-7409.001.patch
>
>
> As discussed on HDFS-7374, we can allow safely allow dead nodes in DECOMMISSION_IN_PROGRESS
to progress to DECOMMISSIONED state if all files on the filesystem are fully-replicated.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message