[GitHub] [hadoop] elek commented on a change in pull request #773: HDDS-1469. Generate default configuration fragments based on annotations

[GitBox <gi...@apache.org>]

elek commented on a change in pull request #773: HDDS-1469. Generate default configuration fragments based on annotations
URL: https://github.com/apache/hadoop/pull/773#discussion_r278874474
 
 

 ##########
 File path: hadoop-hdds/common/src/main/java/org/apache/hadoop/hdds/conf/OzoneConfiguration.java
 ##########
 @@ -161,14 +285,14 @@ public static void activate() {
     Configuration.addDefaultResource("hdfs-default.xml");
     Configuration.addDefaultResource("hdfs-site.xml");
     Configuration.addDefaultResource("ozone-default.xml");
-    Configuration.addDefaultResource("ozone-site.xml");
 
 Review comment:
   We should. But in fact I didn't remove the possibility to use ozone-site.xml I just _moved_ it.
   
   The problem here is that the hadoop Configuration is not designed to be reused. We can use multiple `Configuration.addDefaultResource(String)` but there is no Configuration.addDefaultResource(URL) (and we can't add it as all of the methods are private).
   
   With using configuration fragment we have multiple files with the same name (String) but with different URL (eg. jar://hadoop-ozone-ozone-manager.jar!/ozone-default-generated.xml).
   
   To support this, but _keep_ the same precedence I modified the configuration loading:
   
   ```
   Configuration.addDefaultResource("hdfs-default.xml");
   Configuration.addDefaultResource("hdfs-site.xml");
   Configuration.addDefaultResource("ozone-default.xml");
   
   conf.addResource("jar://ozone-manager.jar!/ozone-site-generated.xml");
   conf.addResource("jar://storage-container-manager.jar!/ozone-site-generated.xml");
   
   conf.addResource("ozone-site.xml");
   ```
   
   The last three lines are in the constructor of OzoneConfiguration and as I would like to load ozone-site.xml at the end I moved it to there.

----------------------------------------------------------------
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:
users@infra.apache.org


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: common-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: common-issues-help@hadoop.apache.org