Ngone51 commented on a change in pull request #35085:
URL: https://github.com/apache/spark/pull/35085#discussion_r830456221
##########
File path:
core/src/main/scala/org/apache/spark/storage/BlockManagerMasterEndpoint.scala
##########
@@ -294,33 +298,74 @@ class BlockManagerMasterEndpoint(
}
}.toSeq
- val removeRddBlockViaExtShuffleServiceFutures =
externalBlockStoreClient.map { shuffleClient =>
- blocksToDeleteByShuffleService.map { case (bmId, blockIds) =>
- Future[Int] {
- val numRemovedBlocks = shuffleClient.removeBlocks(
- bmId.host,
- bmId.port,
- bmId.executorId,
- blockIds.map(_.toString).toArray)
- numRemovedBlocks.get(defaultRpcTimeout.duration.toSeconds,
TimeUnit.SECONDS)
+ val removeRddBlockViaExtShuffleServiceFutures = if
(externalShuffleServiceRddFetchEnabled) {
+ externalBlockStoreClient.map { shuffleClient =>
+ blocksToDeleteByShuffleService.map { case (bmId, blockIds) =>
+ Future[Int] {
+ val numRemovedBlocks = shuffleClient.removeBlocks(
+ bmId.host,
+ bmId.port,
+ bmId.executorId,
+ blockIds.map(_.toString).toArray)
+ numRemovedBlocks.get(defaultRpcTimeout.duration.toSeconds,
TimeUnit.SECONDS)
+ }
}
- }
- }.getOrElse(Seq.empty)
+ }.getOrElse(Seq.empty)
+ } else {
+ Seq.empty
+ }
Future.sequence(removeRddFromExecutorsFutures ++
removeRddBlockViaExtShuffleServiceFutures)
}
private def removeShuffle(shuffleId: Int): Future[Seq[Boolean]] = {
- // Nothing to do in the BlockManagerMasterEndpoint data structures
val removeMsg = RemoveShuffle(shuffleId)
- Future.sequence(
- blockManagerInfo.values.map { bm =>
- bm.storageEndpoint.ask[Boolean](removeMsg).recover {
- // use false as default value means no shuffle data were removed
- handleBlockRemovalFailure("shuffle", shuffleId.toString,
bm.blockManagerId, false)
+ val removeShuffleFromExecutorsFutures = blockManagerInfo.values.map { bm =>
+ bm.storageEndpoint.ask[Boolean](removeMsg).recover {
+ // use false as default value means no shuffle data were removed
+ handleBlockRemovalFailure("shuffle", shuffleId.toString,
bm.blockManagerId, false)
+ }
+ }.toSeq
+
+ // Find all shuffle blocks on executors that are no longer running
Review comment:
Fine. We could do it in a follow-up PR.
> And I'm not sure if all block managers still track shuffles for some
reason even if they don't have any of the shuffle files?
If so, you should also send request to all the external shuffle service..
--
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]
