holdenk commented on a change in pull request #28708:
URL: https://github.com/apache/spark/pull/28708#discussion_r440378160
##########
File path:
core/src/main/scala/org/apache/spark/storage/BlockManagerMasterEndpoint.scala
##########
@@ -334,13 +336,14 @@ class BlockManagerMasterEndpoint(
val info = blockManagerInfo(blockManagerId)
val rddBlocks = info.blocks.keySet().asScala.filter(_.isRDD)
- rddBlocks.map { blockId =>
+ val result = rddBlocks.map { blockId =>
val currentBlockLocations = blockLocations.get(blockId)
val maxReplicas = currentBlockLocations.size + 1
val remainingLocations = currentBlockLocations.toSeq.filter(bm => bm !=
blockManagerId)
val replicateMsg = ReplicateBlock(blockId, remainingLocations,
maxReplicas)
replicateMsg
}.toSeq
+ result
Review comment:
no, I can revert this. I have it in from when I was debugging some stuff
in here.
##########
File path:
core/src/main/scala/org/apache/spark/storage/BlockManagerMasterEndpoint.scala
##########
@@ -489,6 +492,24 @@ class BlockManagerMasterEndpoint(
storageLevel: StorageLevel,
memSize: Long,
diskSize: Long): Boolean = {
+ logInfo(s"Updating block info on master ${blockId} for ${blockManagerId}")
+
+ if (blockId.isInternalShuffle) {
+ blockId match {
+ case ShuffleIndexBlockId(shuffleId, mapId, _) =>
+ // Don't update the map output on just the index block
+ logDebug("Received shuffle index block update for ${shuffleId}
${mapId}, ignoring.")
Review comment:
thanks :)
----------------------------------------------------------------
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
