cloud-fan commented on a change in pull request #24892: [SPARK-25341][Core]
Support rolling back a shuffle map stage and re-generate the shuffle files
URL: https://github.com/apache/spark/pull/24892#discussion_r312916060
##########
File path: core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala
##########
@@ -1602,11 +1627,18 @@ private[spark] class DAGScheduler(
case mapStage: ShuffleMapStage =>
val numMissingPartitions =
mapStage.findMissingPartitions().length
if (numMissingPartitions < mapStage.numTasks) {
- // TODO: support to rollback shuffle files.
- // Currently the shuffle writing is "first write wins",
so we can't re-run a
- // shuffle map stage and overwrite existing shuffle
files. We have to finish
- // SPARK-8029 first.
- abortStage(mapStage, generateErrorMessage(mapStage),
None)
+ if
(sc.getConf.get(config.SHUFFLE_USE_OLD_FETCH_PROTOCOL)) {
+ val reason = "A shuffle map stage with indeterminate
output was failed " +
+ "and retried. However, Spark can only do this while
using the new " +
+ "shuffle block fetching protocol. Please check the
config " +
+ "'spark.shuffle.useOldFetchProtocol', see more
detail in " +
+ "SPARK-27665 and SPARK-25341."
+ abortStage(mapStage, reason, None)
+ } else {
+ logInfo(s"The indeterminate stage $mapStage will be
resubmitted," +
+ " the stage self and all indeterminate parent stage
will be" +
Review comment:
`all indeterminate parent stage` => `all parent stages`. We need to rerun
the parent stages even if they are deterministic.
----------------------------------------------------------------
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.
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]
