[
https://issues.apache.org/jira/browse/HDFS-6244?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14696329#comment-14696329
]
Ming Ma commented on HDFS-6244:
-------------------------------
Thanks [~l201514]!
The patch adds key with prefix of "dfs.federation" to
CommonConfigurationKeysPublic. Not sure if that it is a good place to put it
given federation is specific to HDFS and CommonConfigurationKeysPublic and
Trash are under hadoop-common-project and might be designed to be used by any
FileSystem.
Your early patch had NameNode read the new property defined in hdfs-site.xml
and set the value for {{fs.trash.interval}} before creating {{Trash}}. Any
reason not to go with that?
{{dfs.federation.trash.interval.ns.}} might be misleading as ns might mean
nanosecond. "minutes" might be better. Another thing, maybe we can drop
federation from the name; {{dfs.trash.interval.minutes}} is good enough; just
like how {{dfs.namenode.rpc-address}} is used as prefix for different
namespaces.
It might be useful to add some description for the new property and how it
overrides the {{fs.trash.interval}}.
The patch includes unrelated FairSchedulerPage.
> Make Trash Interval configurable for each of the namespaces
> -----------------------------------------------------------
>
> Key: HDFS-6244
> URL: https://issues.apache.org/jira/browse/HDFS-6244
> Project: Hadoop HDFS
> Issue Type: Improvement
> Affects Versions: 2.0.5-alpha
> Reporter: Siqi Li
> Assignee: Siqi Li
> Labels: BB2015-05-TBR
> Attachments: HDFS-6244.v1.patch, HDFS-6244.v2.patch,
> HDFS-6244.v3.patch, HDFS-6244.v4.patch
>
>
> Somehow we need to avoid the cluster filling up.
> One solution is to have a different trash policy per namespace. However, if
> we can simply make the property configurable per namespace, then the same
> config can be rolled everywhere and we'd be done. This seems simple enough.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
