[
https://issues.apache.org/jira/browse/HDDS-3667?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17192404#comment-17192404
]
Nanda kumar commented on HDDS-3667:
-----------------------------------
[~maobaolong], what advantage do we get by sending a stop signal from the
datanode to SCM? This will require additional logic in Node Management code of
SCM. If there is not much of value addition, we don't have to add additional
complexity in SCM Node Management.
We should not directly go and remove a datanode with an admin command, for
removing a datanode we should do decommissioning. Admin command to remove a
datanode will just confuse the user.
Decommissioining Jira: HDDS-1880
> If we gracefully stop datanode it would be better to notify scm and recon
> --------------------------------------------------------------------------
>
> Key: HDDS-3667
> URL: https://issues.apache.org/jira/browse/HDDS-3667
> Project: Hadoop Distributed Data Store
> Issue Type: New Feature
> Components: Ozone Datanode, Ozone Recon, SCM
> Affects Versions: 1.0.0
> Reporter: maobaolong
> Assignee: Lisheng Sun
> Priority: Minor
> Labels: pull-request-available
>
> if you execute `bin/ozone --daemon stop datanode`, datanode would be better
> if it handle the signal and the stop dn request to scm, then scm update
> datanode state to dead.
> It would be better if you provide a admin cli tool to support remove datanode
> manually.
> {code:bash}
> ozone admin datanode remove <UUID>
> {code}
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: ozone-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: ozone-issues-h...@hadoop.apache.org
