[
https://issues.apache.org/jira/browse/HDFS-13767?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16564011#comment-16564011
]
Chen Liang commented on HDFS-13767:
-----------------------------------
Thanks for the followup discussion [~xkrogen]! Many thanks to the sharing about
the locking overhead. Also it is an interesting case I did not think about
earlier, so thanks for pointing out. But I would say it's on a slightly
different line, as it is about ANN returning the correct transaction id. I
think we do need to take care of this, but we can always make an check that if
current state is ANN, we return the correct version, but not necessarily for
Observer. This being said, given what you shared about the locking overhead, I
will switch to getting correct id in the next patch, and make a note in
comments so that when working on general optimization later, we don't forget
about this discussion.
> Add msync server implementation.
> --------------------------------
>
> Key: HDFS-13767
> URL: https://issues.apache.org/jira/browse/HDFS-13767
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: namenode
> Reporter: Chen Liang
> Assignee: Chen Liang
> Priority: Major
> Attachments: HDFS-13767.WIP.001.patch, HDFS-13767.WIP.002.patch,
> HDFS-13767.WIP.003.patch
>
>
> This is a followup on HDFS-13688, where msync API is introduced to
> {{ClientProtocol}} but the server side implementation is missing. This is
> Jira is to implement the server side logic.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
