hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Tsz Wo (Nicholas), SZE (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HDFS-2108) Move datanode heartbeat handling to BlockManager
Date Tue, 02 Aug 2011 23:26:27 GMT

     [ https://issues.apache.org/jira/browse/HDFS-2108?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Tsz Wo (Nicholas), SZE updated HDFS-2108:
-----------------------------------------

    Attachment: h2108_20110802.patch

Hi Suresh, thanks for the comments.

h2108_20110802.patch: see below.

1, 3, 4.
done.

2. Why have heartbeatManager in BlockManager. Why not just DatanodeManager to removeDataNode,
stop and start decommission etc.?

Renamed BlockManager.removeDatanode(..) to removeBlocksAssociatedTo(..)
and moved decommission related methods to DatanodeManager.

> As much as possible, we should avoid getting HeartbeatManager from the DatanodeManager.

Added an interface DatanodeStatistics so that HeartbeatManager can be changed from public
to package private.

> Move datanode heartbeat handling to BlockManager
> ------------------------------------------------
>
>                 Key: HDFS-2108
>                 URL: https://issues.apache.org/jira/browse/HDFS-2108
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>          Components: name-node
>            Reporter: Tsz Wo (Nicholas), SZE
>            Assignee: Tsz Wo (Nicholas), SZE
>         Attachments: h2108_20110727.patch, h2108_20110727b.patch, h2108_20110801.patch,
h2108_20110802.patch
>
>
> Logically, datanodes should heartbeat to block manager instead of name system.  Therefore,
we should move datanode heartbeat handling code to {{BlockManager}}.

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

        

Mime
View raw message