[
https://issues.apache.org/jira/browse/HDFS-11279?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15795742#comment-15795742
]
Xiaoyu Yao commented on HDFS-11279:
-----------------------------------
Thanks [~hanishakoneru] for working on this. The patch looks pretty good to me.
I just have two NITs, +1 otherwise.
Datanode.java
NIT: Line50, import can be moved down to follow the alphabet order
importorg.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsVolumeImpl;
NIT: Line 2470, can you add @VisibleForTesting for DataNode#getVolume?
Or I would suggest moving DataNode#getVolume(File basePath) into a test utility
class such as DataNodeTestUtils#getVolume
{code}
Public FsVolumeImpl getVolume(DataNode dn,File basePath)
{code}
> Cleanup unused DataNode#checkDiskErrorAsync()
> ---------------------------------------------
>
> Key: HDFS-11279
> URL: https://issues.apache.org/jira/browse/HDFS-11279
> Project: Hadoop HDFS
> Issue Type: Improvement
> Reporter: Xiaoyu Yao
> Assignee: Hanisha Koneru
> Priority: Minor
> Attachments: HDFS-11279.000.patch
>
>
> After HDFS-11274, we will not trigger checking all datanode volumes upon IO
> failure on a single volume. This makes the original implementation
> DataNode#checkDiskErrorAsync and DatasetVolumeChecker#checkAllVolumesAsync()
> not used in any of the production code.
> This ticket is opened to remove these unused code and related tests if any.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
