[
https://issues.apache.org/jira/browse/HDFS-7722?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14353605#comment-14353605
]
Chris Nauroth commented on HDFS-7722:
-------------------------------------
Eddy, it looks good. I have just one minor nit. In
{{TestDataNodeVolumeFailureReporting}}, please remove the commented out lines
of test code for the final version of the patch. Also, we can no longer remove
the import of {{org.apache.hadoop.hdfs.protocol.Block}}, because another patch
started using it recently.
bq. I suggest to have a following JIRA...
Please feel free to do that if you wish, but I actually don't think it's
necessary. In general, I don't expect permanent removal of a volume to be the
typical recovery procedure. Instead, I expect a more typical recovery
procedure to be like you described: replace the faulty disk. Since that works
fine, I think it would be overkill at this point to put in dedicated
functionality to cover something that is probably a very rare edge case in
practical deployments.
Thanks for working on this!
> DataNode#checkDiskError should also remove Storage when error is found.
> -----------------------------------------------------------------------
>
> Key: HDFS-7722
> URL: https://issues.apache.org/jira/browse/HDFS-7722
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: datanode
> Affects Versions: 2.6.0
> Reporter: Lei (Eddy) Xu
> Assignee: Lei (Eddy) Xu
> Attachments: HDFS-7722.000.patch, HDFS-7722.001.patch,
> HDFS-7722.002.patch
>
>
> When {{DataNode#checkDiskError}} found disk errors, it removes all block
> metadatas from {{FsDatasetImpl}}. However, it does not removed the
> corresponding {{DataStorage}} and {{BlockPoolSliceStorage}}.
> The result is that, we could not directly run {{reconfig}} to hot swap the
> failure disks without changing the configure file.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
