[
https://issues.apache.org/jira/browse/HDFS-1547?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12980497#action_12980497
]
Todd Lipcon commented on HDFS-1547:
-----------------------------------
One potential issue, not sure if it's realistically a problem: In
BlockPlacementPolicyDefault, the average load on the cluster is calculated by
dividing FSN.totalLoad by the number of nodes. Since totalLoad no longer
includes decomissioning nodes, I wonder whether this could create a situation
where *all* nodes are considered "overloaded" and we can't allocate new blocks.
> Improve decommission mechanism
> ------------------------------
>
> Key: HDFS-1547
> URL: https://issues.apache.org/jira/browse/HDFS-1547
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: name-node
> Affects Versions: 0.23.0
> Reporter: Suresh Srinivas
> Assignee: Suresh Srinivas
> Fix For: 0.23.0
>
> Attachments: HDFS-1547.1.patch, HDFS-1547.patch
>
>
> Current decommission mechanism driven using exclude file has several issues.
> This bug proposes some changes in the mechanism for better manageability. See
> the proposal in the next comment for more details.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
