[
https://issues.apache.org/jira/browse/HDFS-6958?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14113961#comment-14113961
]
Lei (Eddy) Xu commented on HDFS-6958:
-------------------------------------
[~atm] These two failures are all related to balancer, which is not related to
this patch.
> Fix the condition of reporting failures in BlockPoolSlice
> ---------------------------------------------------------
>
> Key: HDFS-6958
> URL: https://issues.apache.org/jira/browse/HDFS-6958
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: datanode
> Affects Versions: 2.5.0
> Reporter: Lei (Eddy) Xu
> Assignee: Lei (Eddy) Xu
> Priority: Trivial
> Attachments: HDFS-6958.000.patch
>
>
> {{BlockPoolSlice#addToReplicsMap()}} should log the failure for the operation
> that deletes restart metadata file.
> The current code did the opposite, log the failures on succeed deletion:
> {code:java}
> if (restartMeta.delete()) {
> FsDatasetImpl.LOG.warn("Failed to delete restart meta file: " +
> restartMeta.getPath());
> }
> {code}
--
This message was sent by Atlassian JIRA
(v6.2#6252)
