vanzin commented on a change in pull request #23599: [SPARK-24793][K8s] Enhance
spark-submit for app management
URL: https://github.com/apache/spark/pull/23599#discussion_r256538217
##########
File path: core/src/main/scala/org/apache/spark/deploy/SparkSubmit.scala
##########
@@ -1358,3 +1387,12 @@ private case class OptionAssigner(
deployMode: Int,
clOption: String = null,
confKey: String = null)
+
+private[spark] trait SparkSubmitOperationsClient {
+
+ def kill(args: SparkSubmitArguments): Unit
Review comment:
There are a bunch of configuration options that can affect how Spark talks
to the k8s server. Wouldn't it be better to provide a `SparkConf` to these
methods instead of forcing them to figure out how to get that data?
You have code in your k8s implementation that basically duplicates code that
already exists in `SparkSubmit` and could be reused with just a few changes.
It's also a little weird to provide `SparkSubmitArguments` to the plugins,
since it's a mutable class that exposes a lot of unrelated stuff. I think
`kill(submissionId: String, conf: SparkConf)` would be a better interface, but
not sure whether that's enough.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
