[
https://issues.apache.org/jira/browse/HDFS-5662?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Brandon Li updated HDFS-5662:
-----------------------------
Attachment: HDFS-5662.branch2.3.patch
The patch to branch2.3 is slightly different in the unit test. Attached it here.
> Can't decommission a DataNode due to file's replication factor larger than
> the rest of the cluster size
> -------------------------------------------------------------------------------------------------------
>
> Key: HDFS-5662
> URL: https://issues.apache.org/jira/browse/HDFS-5662
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: namenode
> Reporter: Brandon Li
> Assignee: Brandon Li
> Fix For: 2.3.0
>
> Attachments: HDFS-5662.001.patch, HDFS-5662.002.patch,
> HDFS-5662.branch2.3.patch
>
>
> A datanode can't be decommissioned if it has replica that belongs to a file
> with a replication factor larger than the rest of the cluster size.
> One way to fix this is to have some kind of minimum replication factor
> setting and thus any datanode can be decommissioned regardless of the largest
> replication factor it's related to.
--
This message was sent by Atlassian JIRA
(v6.1.4#6159)
