[
https://issues.apache.org/jira/browse/FLINK-22257?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17364762#comment-17364762
]
Timo Walther commented on FLINK-22257:
--------------------------------------
Maybe but not always. We should think about the layers or teams that usually
handle different stages of Flink application lifecycle. Maybe something like
CLUSTER, JOB, PIPELINE. But this is just initial idea, it requires deeper
investigation.
> Clarify Flink ConfigOptions Usage
> ---------------------------------
>
> Key: FLINK-22257
> URL: https://issues.apache.org/jira/browse/FLINK-22257
> Project: Flink
> Issue Type: Improvement
> Components: Runtime / Configuration
> Reporter: Fabian Paul
> Priority: Minor
>
> For users, it is hard to determine which ConfigOptions are relevant for the
> different stages of a Flink application.
> Beginning from the translation of the user program to the execution on the
> cluster. In particular which options can be configured through the different
> channels.
> * Cluster configuration (i.e. flink-conf.yaml)
> * Application configuration, code-based
>
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
