Github user sryza commented on a diff in the pull request:
https://github.com/apache/spark/pull/4020#discussion_r23057216
--- Diff: core/src/main/scala/org/apache/spark/executor/Executor.scala ---
@@ -257,8 +257,8 @@ private[spark] class Executor(
val serviceTime = System.currentTimeMillis() - taskStart
val metrics = attemptedTask.flatMap(t => t.metrics)
for (m <- metrics) {
- m.executorRunTime = serviceTime
- m.jvmGCTime = gcTime - startGCTime
+ m.incExecutorRunTime(serviceTime)
--- End diff --
Task GC time is measured by looking at ((total time spent in GC since
process started) - (total time spent in GC since process started, snapshotted
when task started)). Incrementing every time we update the metric would end up
double-counting certain time regions. A few other metrics work this way as
well. If you do notice issues with the logic, it might be good to propose
fixes in a separate JIRA, so that we can give them careful consideration on
their own.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
