hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Weiwei Yang (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-11773) Ozone: KSM : add listVolumes
Date Mon, 26 Jun 2017 10:28:00 GMT

    [ https://issues.apache.org/jira/browse/HDFS-11773?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16062886#comment-16062886
] 

Weiwei Yang commented on HDFS-11773:
------------------------------------

Patch uploaded, v1 patch implemented listVolumes in KSM. Unit tests can be found in {{TestKeySpaceManager#testListVolumes}}
and real cluster tests can be found in attachment [^listVolume_tests.log], please help to
review.

Note, this patch doesn't fix following issues
# ozone CLI listVolume command doesn't support prefix, startKey and maxKey arguments
# the permission check module right now only makes sure only root user can scan all volumes.
But it doesn't check if userA can listVolumes owned by userB
# listVolume result has some missing fields, because those info isn't updated to KSM db yet.
See more in the todo item in {{DistributedStorageHandler#listVolumes}}

I will create JIRAs as following up tasks for them, to avoid we stuck with a big patch.

> Ozone: KSM : add listVolumes
> ----------------------------
>
>                 Key: HDFS-11773
>                 URL: https://issues.apache.org/jira/browse/HDFS-11773
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>          Components: ozone
>    Affects Versions: HDFS-7240
>            Reporter: Anu Engineer
>            Assignee: Weiwei Yang
>         Attachments: HDFS-11773-HDFS-7240.001.patch, listVolume_tests.log
>
>
> List volume call can be used in three different contexts. One is for the administrators
to list all volumes in a cluster. Second is for the administrator to list the volumes owned
by a specific user. Third is a user listing the volumes owned by himself/herself.
> Since these calls can return large number of entries the rest protocol supports paging.
Paging is supported by the use of prevKey, prefix and maxKeys. The caller is aware the this
call is neither atomic nor consistent. So we can iterate over the list even while changes
are happening to the list.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-help@hadoop.apache.org


Mime
View raw message