[
https://issues.apache.org/jira/browse/FLINK-16005?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17040813#comment-17040813
]
Till Rohrmann commented on FLINK-16005:
---------------------------------------
I don't fully understand the last problem [~fly_in_gis]. My understanding was
that there would be one single place where we load the {{YarnConfiguration}}
and override the keys from the Flink {{Configuration}}. How would that not work
with the code snippet you've posted?
Before diving into the implementation [~ZhenqiuHuang] could you explain the
exact implementation plan? Please also take {{HadoopConfigLoader}} into account
and properly define what we wanna support (e.g. allowing only to set yarn
configuration options vs. hadoop and yarn). I guess it would make sense to take
a look at the {{YarnConfiguration}} instantiation sites and see how we can
unify them.
> Propagate yarn.application.classpath from client to TaskManager Classpath
> -------------------------------------------------------------------------
>
> Key: FLINK-16005
> URL: https://issues.apache.org/jira/browse/FLINK-16005
> Project: Flink
> Issue Type: Improvement
> Components: Deployment / YARN
> Reporter: Zhenqiu Huang
> Priority: Major
>
> When Flink users want to override the hadoop yarn container classpath, they
> should just specify the yarn.application.classpath in yarn-site.xml from cli
> side. But currently, the classpath setting can only be used in flink
> application master, the classpath of TM is still determined by the setting in
> yarn host.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
