Ngone51 commented on a change in pull request #28924:
URL: https://github.com/apache/spark/pull/28924#discussion_r448951365
##########
File path:
core/src/main/scala/org/apache/spark/storage/BlockManagerMasterEndpoint.scala
##########
@@ -235,7 +269,10 @@ class BlockManagerMasterEndpoint(
val removeMsg = RemoveShuffle(shuffleId)
Future.sequence(
blockManagerInfo.values.map { bm =>
- bm.slaveEndpoint.ask[Boolean](removeMsg)
+ bm.slaveEndpoint.ask[Boolean](removeMsg).recover {
+ // use false as default value means no shuffle data were removed
+ handleFailure("shuffle", shuffleId.toString, bm.blockManagerId,
false)
Review comment:
I think we miss it previously. The failure of removing blocks should be
considered as non-fatal. For example, it's not worth failing the whole
application if we fail to remove some blocks just after some heavy computation.
----------------------------------------------------------------
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]
