Github user liyezhang556520 commented on a diff in the pull request:
https://github.com/apache/spark/pull/2792#discussion_r18997684
--- Diff:
core/src/main/scala/org/apache/spark/storage/BlockManagerMasterActor.scala ---
@@ -457,16 +457,18 @@ private[spark] class BlockManagerInfo(
if (_blocks.containsKey(blockId)) {
// The block exists on the slave already.
- val originalLevel: StorageLevel = _blocks.get(blockId).storageLevel
+ val blockStatus: BlockStatus = _blocks.get(blockId)
+ val originalLevel: StorageLevel = blockStatus.storageLevel
+ val originalMemSize: Long = blockStatus.memSize
if (originalLevel.useMemory) {
- _remainingMem += memSize
+ _remainingMem += originalMemSize
--- End diff --
"memSize" is the current memory that the block consume, "originalMemSize"
is the memory used before block updating. These two size might be not the same.
So here, if the block exists on the slave already, `_remainingmem` should first
`_remainingMem += originalMemSize` and then `_remainingMem -= memSize` if the
new storage level is "useMemory". Do I have misunderstanding of these two mem
size?
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
