Github user tgravescs commented on a diff in the pull request: https://github.com/apache/spark/pull/2311#discussion_r17882325 --- Diff: yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ApplicationMaster.scala --- @@ -91,7 +94,11 @@ private[spark] class ApplicationMaster(args: ApplicationMasterArguments, if (sc != null) { logInfo("Invoking sc stop from shutdown hook") sc.stop() - finish(FinalApplicationStatus.SUCCEEDED) + } + + // Shuts down the AM. + if (!finished) { --- End diff -- I think this could cause problems with YARN retry logic. We shouldn't be finishing (unregistering with RM) unless we've explicitly succeeded or failed and don't want a retry to happen in those weird cases. Since this is in shutdown hook I'm not sure we explicitly know that. i guess it comes down to us defining when the spark app is cleanly exiting. This could be failed, killed, or succeeded. Unfortunately I'm not sure spark really has this final reporting.
--- 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 infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org