csurong opened a new pull request, #4302:
URL: https://github.com/apache/amoro/pull/4302

   ## Why are the changes needed?
   
   `KUBERNETES_EXECUTOR_LABEL_PREFIX` currently uses the driver label prefix. 
Because Spark options are stored by key in a map, the intended executor entries 
overwrite the same driver keys and no `spark.kubernetes.executor.label.*` 
options are emitted. Executor pods therefore miss Amoro's optimizer labels for 
observability and label-based operations.
   
   Closes #4301.
   
   ## Brief change log
   
   - Use `spark.kubernetes.executor.label.` for executor pod labels.
   - Add regression coverage for all optimizer labels on both driver and 
executor startup arguments.
   
   ## How was this patch tested?
   
   - [x] Added a regression test that fails before the prefix correction and 
passes after it.
   - [x] `mvn test -pl amoro-ams -am -Dtest=TestSparkOptimizerContainer 
-Dsurefire.failIfNoSpecifiedTests=false -Pskip-dashboard-build` (JDK 11; 5 
tests passed).
   - [x] `mvn validate -pl amoro-ams` (JDK 11; Spotless and Checkstyle passed).
   - [ ] Screenshots for manual tests (not applicable; no UI change).
   
   ## Documentation
   
   - Does this pull request introduce a new feature? no
   - If yes, how is the feature documented? not applicable


-- 
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.

To unsubscribe, e-mail: [email protected]

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

Reply via email to