You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@flink.apache.org by "Xintong Song (Jira)" <ji...@apache.org> on 2021/03/26 02:56:00 UTC

[jira] [Commented] (FLINK-21982) jobs should use client org.apache.hadoop.config on yarn

    [ https://issues.apache.org/jira/browse/FLINK-21982?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17309112#comment-17309112 ] 

Xintong Song commented on FLINK-21982:
--------------------------------------

Hi [~Bo Cui],
I'm trying to understand, why would a job use the hadoop configurations from the client side rather than those on the Yarn cluster?

> jobs should use client org.apache.hadoop.config on yarn
> -------------------------------------------------------
>
>                 Key: FLINK-21982
>                 URL: https://issues.apache.org/jira/browse/FLINK-21982
>             Project: Flink
>          Issue Type: Bug
>          Components: Deployment / YARN
>    Affects Versions: 1.13.0
>            Reporter: Bo Cui
>            Priority: Major
>
> currently, jobs use the Yarn server configuration during execution. 
> i think we should submit the configuration to the HDFS, like MR job.xml...etc
> and container fetches the job.xml and creates a soft link(hdfs-site core-site yarn-site) during container initializes
> and then the configuration can obtain these resources through classLoader.getResource(String) during configuration initializes



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