[
https://issues.apache.org/jira/browse/HDFS-9392?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ming Ma updated HDFS-9392:
--------------------------
Attachment: HDFS-9392.patch
Here is the draft patch. It also includes datanode state machine transition
under some scenarios and part of HDFS-9388 to reuse test code. The complete
state machine transition such as ENTERING_MAINTENANCE to IN_MAINTENANCE and the
block management will be handled by HDFS-9390.
* Add maintenance support to the json based configuration.
* If the DN is {{DECOMMISSIONED}} or {{DEAD}}, transition it to
{{IN_MAINTENANCE}}.
* If the DN is {{NORMAL, Live}} or {{DECOMMISSION_IN_PROGRESS, Live}},
transition it to {{ENTERING_MAINTENANCE, Live}}.
* Upon maintenance expiration, transition it to {{NORMAL}} state.
> Admins support for maintenance state
> ------------------------------------
>
> Key: HDFS-9392
> URL: https://issues.apache.org/jira/browse/HDFS-9392
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: Ming Ma
> Attachments: HDFS-9392.patch
>
>
> This is to allow admins to put nodes into maintenance state with optional
> timeout value as well as take nodes out of maintenance state. Likely we will
> leverage what we come up in https://issues.apache.org/jira/browse/HDFS-9005.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
