[
https://issues.apache.org/jira/browse/HDFS-17201?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17766734#comment-17766734
]
farmmamba commented on HDFS-17201:
----------------------------------
[~hexiaoqiao] [~tomscut] Hi, sir. Sorry for disturbing you here. I have a
question here and needs your help.
Firstly, take method FsDatasetImpl#contains as example, this method only
acquire BLOCK_POOl read lock, but there exists logic that get ReplicaInfo from
volumeMap.
Some other methods also have such usages.
Secondly, my question is why we just use BLOCK_POOl read lock here rather than
VOLUME read lock ? Do we better change those BLOCK_POOl read locks to VOLUME
read locks?
Look forward to your reply. Thanks a lot.
> some methods in FsDatasetImpl should acquire readLock with LockLevel.VOLUME
> ---------------------------------------------------------------------------
>
> Key: HDFS-17201
> URL: https://issues.apache.org/jira/browse/HDFS-17201
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: datanode
> Reporter: farmmamba
> Assignee: farmmamba
> Priority: Major
> Fix For: 3.4.0
>
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
