[
https://issues.apache.org/jira/browse/HDFS-12083?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16086818#comment-16086818
]
Yiqun Lin commented on HDFS-12083:
----------------------------------
Thanks [~nandakumar131] for the comment, your comment makes sense to me. I'm +1
on the latest patch.
[~vagarychen], In the mainly use cases, list operation is used for multi-page
view. The previous key is got from previous list result. And then use this
previous key to get next page list. This behaviour should be a right way. For
this usage way, we need to exclude the previous key.
bq. how about passing a new boolean flag to indicate whether including start
key or not?
I think this can be a improvement work.
There are no test failure. This patch is a important change, I will commit this
at the end of day in case [~cheersyang] has other comments. Thanks.
> Ozone: KSM: previous key has to be excluded from result in listVolumes,
> listBuckets and listKeys
> ------------------------------------------------------------------------------------------------
>
> Key: HDFS-12083
> URL: https://issues.apache.org/jira/browse/HDFS-12083
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: ozone
> Reporter: Nandakumar
> Assignee: Nandakumar
> Priority: Critical
> Attachments: HDFS-12083-HDFS-7240.000.patch,
> HDFS-12083-HDFS-7240.001.patch
>
>
> When previous key is set as part of list calls [listVolume, listBuckets &
> listKeys], the result includes previous key, there is no need to have this in
> the result.
> Since previous key is present as part of result, we will never receive an
> empty list in the subsequent list calls, this makes it difficult to have a
> exit criteria where we want to get all the values using multiple list calls
> (with previous-key set).
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
