cloud-fan commented on code in PR #36564:
URL: https://github.com/apache/spark/pull/36564#discussion_r892511172
##########
core/src/main/scala/org/apache/spark/mapred/SparkHadoopMapRedUtil.scala:
##########
@@ -76,6 +76,8 @@ object SparkHadoopMapRedUtil extends Logging {
if (canCommit) {
performCommit()
+ outputCommitCoordinator.commitSuccess(ctx.stageId(),
Review Comment:
It's possible that the executor gets killed right after `performCommit()`
and we still have the data duplication issue. I don't think we need this
`commit success` event. We should mark the partition as committed once
`canCommit` returns true.
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
