[
https://issues.apache.org/jira/browse/HDFS-17269?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17803512#comment-17803512
]
ASF GitHub Bot commented on HDFS-17269:
---------------------------------------
slfan1989 commented on PR #6312:
URL: https://github.com/apache/hadoop/pull/6312#issuecomment-1878479567
> @LiuGuH Would it be more chaotic to display the trash directories of all
clusters at once? Should we allow users to choose whether to display the trash
directories of all subgroups through more configuration options, instead of
enforcing such behavior
I agree with @zhtttylz idea. For a cluster with 7-8 NS, it is not friendly.
> RBF: Listing trash directory should return subdirs from all subclusters.
> ------------------------------------------------------------------------
>
> Key: HDFS-17269
> URL: https://issues.apache.org/jira/browse/HDFS-17269
> Project: Hadoop HDFS
> Issue Type: Improvement
> Reporter: liuguanghua
> Assignee: liuguanghua
> Priority: Minor
> Labels: pull-request-available
>
> Same scenario as HDFS-17263
> If user trash config fs.trash.checkpoint.interval set to 10min in namenodes,
> the trash root dir /user/$USER/.Trash/Current will be very 10 min renamed to
> /user/$USER/.Trash/timestamp .
>
> When user ls /user/$USER/.Trash, it should be return blow:
> /user/$USER/.Trash/Current
> /user/$USER/.Trash/timestamp (This is invisible now)
>
> So we should make that user ls trash root dir can see all trash subdirs in
> all nameservices which user has any mountpoint in nameservice.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
