You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Wenlong Lyu (Jira)" <ji...@apache.org> on 2020/04/15 09:31:00 UTC

[jira] [Commented] (FLINK-17110) Make StreamExecutionEnvironment#configure also affects StreamExecutionEnvironment#configuration

    [ https://issues.apache.org/jira/browse/FLINK-17110?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17083957#comment-17083957 ] 

Wenlong Lyu commented on FLINK-17110:
-------------------------------------

hi, [~aljoscha][~kkl0u] Do you have any more comments? Here is a patch : https://github.com/wenlong88/flink/commit/651a48ce4d837623e6eacb4814bd1e43c686fdd4, mainly contains two points of change:
1. use PipelineExecutorUtils to generate JobGraph so that the configuration in StreamExecutionEnvironment can affect on JobGraph, unifying the manner on RemoteExecutor, LocalExecutor, ClasspathJobGraphRetriever and web submit handler.
2. make configuration in StreamExecutionEnvironment configurable by StreamExecutionEnvironment#configure, so that user can update the configuration in the program.

> Make StreamExecutionEnvironment#configure also affects StreamExecutionEnvironment#configuration
> -----------------------------------------------------------------------------------------------
>
>                 Key: FLINK-17110
>                 URL: https://issues.apache.org/jira/browse/FLINK-17110
>             Project: Flink
>          Issue Type: Improvement
>            Reporter: Wenlong Lyu
>            Priority: Major
>
> If StreamExecutionEnvironment#configure can also affect the configuration in StreamExecutionEnvironment, we can easily not only add some library jars or classpaths dynamically according to the job we want to run which is quite important for a platform product, but also optimize some runtime configuration in program.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)