[
https://issues.apache.org/jira/browse/FLINK-22585?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Xintong Song updated FLINK-22585:
---------------------------------
Fix Version/s: (was: 1.14.0)
1.15.0
> Add deprecated message when "-m yarn-cluster" is used
> -----------------------------------------------------
>
> Key: FLINK-22585
> URL: https://issues.apache.org/jira/browse/FLINK-22585
> Project: Flink
> Issue Type: Improvement
> Components: Deployment / YARN
> Reporter: Yang Wang
> Assignee: Rainie Li
> Priority: Major
> Labels: auto-unassigned, pull-request-available, stale-assigned
> Fix For: 1.15.0
>
> Time Spent: 4h
> Remaining Estimate: 0h
>
> The unified executor interface has been introduced for a long time, which
> could be activated by "\--target
> yarn-per-job/yarn-session/kubernetes-application". It is more descriptive and
> clearer. We should try to deprecate "-m yarn-cluster" and suggest our users
> to use the new CLI commands.
>
> However, AFAIK, many companies are using some CLI commands to integrate with
> their deployers. So we could not remove the "-m yarn-cluster" very soon.
> Maybe we could do it in the release 2.0 since we could do some breaking
> changes.
>
> For now, I suggest to add the {{@Deprecated}} annotation and printing a WARN
> log message when "-m yarn-cluster" is used. It is useful to let the users
> know the long-term goals and migrate asap.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
