[
https://issues.apache.org/jira/browse/HDFS-8688?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14630723#comment-14630723
]
Walter Su commented on HDFS-8688:
---------------------------------
By the way, the patch changes the logic in the special case as you said before.
The change/patch is valid.
Assume live nodes=0. It must be in needReplications or pendingReplications.
It's ok blockHasEnoughRacks() return true ( and only if there is only one rack
in cluster. The returned value is meaningless in this case)
Assume one replica got copied from decommissioned nodes, Now live nodes=1. The
logic is the same as before the patch. (Both return true, Because they both
check hasClusterEverBeenMultiRack() )
> replace shouldCheckForEnoughRacks with hasClusterEverBeenMultiRack
> ------------------------------------------------------------------
>
> Key: HDFS-8688
> URL: https://issues.apache.org/jira/browse/HDFS-8688
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: Walter Su
> Assignee: Walter Su
> Attachments: HDFS-8688.01.patch, HDFS-8688.02.patch
>
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
