[GitHub] [flink] yuruguo commented on a change in pull request #14477: [FLINK-16005][yarn] Support yarn and hadoop config override

Thu, 07 Jan 2021 02:37:40 -0800 


yuruguo commented on a change in pull request #14477:
URL: https://github.com/apache/flink/pull/14477#discussion_r553245985



##########
File path: 
flink-yarn/src/main/java/org/apache/flink/yarn/YarnClusterClientFactory.java
##########
@@ -77,6 +78,8 @@ public ApplicationId getClusterId(Configuration 
configuration) {
     private YarnClusterDescriptor getClusterDescriptor(Configuration 
configuration) {
         final YarnClient yarnClient = YarnClient.createYarnClient();
         final YarnConfiguration yarnConfiguration = new YarnConfiguration();
+        
yarnConfiguration.addResource(HadoopUtils.getHadoopConfiguration(configuration));

Review comment:
       I find this `yarnConfiguration` can be used to get FileSystem in 
YarnClusterDescriptor.class, such as in`startAppMaster` method:
   
![image](https://user-images.githubusercontent.com/13013780/103879642-172d3800-5113-11eb-860e-fe5f1b7bbc45.png)
   So add `HadoopConfiguration` to benefit the filesystem.




----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to