Github user squito commented on a diff in the pull request:
https://github.com/apache/spark/pull/22612#discussion_r234347780
--- Diff:
core/src/main/scala/org/apache/spark/metrics/ExecutorMetricType.scala ---
@@ -84,6 +122,8 @@ case object MappedPoolMemory extends
MBeanExecutorMetricType(
"java.nio:type=BufferPool,name=mapped")
private[spark] object ExecutorMetricType {
+ final val pTreeInfo = new ProcfsBasedSystems
--- End diff --
this is a weird place to keep this, unless there is some really good reason
for it. I think it should go inside `ProcessTreeMetrics`.
also I'm not sure what the problem was with making it an object. Seems to
work for me. its a bit different now as there are arguments to the constructor
for testing -- but you could still have an object which extends the class
```scala
private[spark] object ProcfsBasedSystems extends
ProcfsBasedSystems("/proc/")
```
though doesn't really seem to have much value.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
