[
https://issues.apache.org/jira/browse/HDFS-17129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17807034#comment-17807034
]
ASF GitHub Bot commented on HDFS-17129:
---------------------------------------
tasanuma commented on PR #6244:
URL: https://github.com/apache/hadoop/pull/6244#issuecomment-1893063350
I'd like to share our situation in more detail. In my company, we are using
hadoop-3.3.0 with many patches. We had a problem with IBR becoming very slow,
which is exactly the issue described in HDFS-16898. Then, we backported
HDFS-16061 and the problem was solved (HDFS-16898 didn't exist at that time).
Now, I have confirmed that our problem is fixed by HDFS-16898 without
HDFS-16061. So, I would suggest only reverting HDFS-16061 for now. It may no
longer be needed.
> mis-order of ibr and fbr on datanode
> -------------------------------------
>
> Key: HDFS-17129
> URL: https://issues.apache.org/jira/browse/HDFS-17129
> Project: Hadoop HDFS
> Issue Type: Bug
> Affects Versions: 3.4.0, 3.3.9, 3.3.6
> Environment: hdfs3.4.0
> Reporter: liuguanghua
> Assignee: liuguanghua
> Priority: Blocker
> Labels: pull-request-available
>
> HDFS-16016 , provide new thread to handler IBR. That is a greate improvement.
> But it maybe casue the mis-order of ibr and fbr
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
