Github user mridulm commented on a diff in the pull request:
https://github.com/apache/spark/pull/17295#discussion_r106779457
--- Diff: core/src/main/scala/org/apache/spark/storage/DiskStore.scala ---
@@ -73,55 +86,219 @@ private[spark] class DiskStore(conf: SparkConf,
diskManager: DiskBlockManager) e
}
def putBytes(blockId: BlockId, bytes: ChunkedByteBuffer): Unit = {
- put(blockId) { fileOutputStream =>
- val channel = fileOutputStream.getChannel
- Utils.tryWithSafeFinally {
- bytes.writeFully(channel)
- } {
- channel.close()
- }
+ put(blockId) { channel =>
+ bytes.writeFully(channel)
}
}
- def getBytes(blockId: BlockId): ChunkedByteBuffer = {
+ def getBytes(blockId: BlockId): BlockData = {
val file = diskManager.getFile(blockId.name)
- val channel = new RandomAccessFile(file, "r").getChannel
- Utils.tryWithSafeFinally {
- // For small files, directly read rather than memory map
- if (file.length < minMemoryMapBytes) {
- val buf = ByteBuffer.allocate(file.length.toInt)
- channel.position(0)
- while (buf.remaining() != 0) {
- if (channel.read(buf) == -1) {
- throw new IOException("Reached EOF before filling buffer\n" +
-
s"offset=0\nfile=${file.getAbsolutePath}\nbuf.remaining=${buf.remaining}")
+ val blockSize = getSize(blockId)
+
+ securityManager.getIOEncryptionKey() match {
+ case Some(key) =>
+ // Encrypted blocks cannot be memory mapped; return a special
object that does decryption
+ // and provides InputStream / FileRegion implementations for
reading the data.
+ new EncryptedBlockData(file, blockSize, conf, key)
+
+ case _ =>
+ val channel = new FileInputStream(file).getChannel()
+ if (blockSize < minMemoryMapBytes) {
+ // For small files, directly read rather than memory map.
+ Utils.tryWithSafeFinally {
+ val buf = ByteBuffer.allocate(blockSize.toInt)
+ while (buf.remaining() > 0) {
+ channel.read(buf)
+ }
+ buf.flip()
+ new ByteBufferBlockData(new ChunkedByteBuffer(buf))
+ } {
+ channel.close()
+ }
+ } else {
+ Utils.tryWithSafeFinally {
+ new ByteBufferBlockData(
+ new ChunkedByteBuffer(channel.map(MapMode.READ_ONLY, 0,
file.length)))
+ } {
+ channel.close()
}
}
- buf.flip()
- new ChunkedByteBuffer(buf)
- } else {
- new ChunkedByteBuffer(channel.map(MapMode.READ_ONLY, 0,
file.length))
- }
- } {
- channel.close()
}
}
def remove(blockId: BlockId): Boolean = {
val file = diskManager.getFile(blockId.name)
- if (file.exists()) {
- val ret = file.delete()
- if (!ret) {
- logWarning(s"Error deleting ${file.getPath()}")
+ val meta = diskManager.getMetadataFile(blockId)
+
+ def delete(f: File): Boolean = {
+ if (f.exists()) {
+ val ret = f.delete()
+ if (!ret) {
+ logWarning(s"Error deleting ${file.getPath()}")
+ }
+
+ ret
+ } else {
+ false
}
- ret
- } else {
- false
}
+
+ delete(file) & delete(meta)
}
def contains(blockId: BlockId): Boolean = {
val file = diskManager.getFile(blockId.name)
file.exists()
}
+
+ private def openForWrite(file: File): WritableByteChannel = {
+ val out = new FileOutputStream(file).getChannel()
+ try {
+ securityManager.getIOEncryptionKey().map { key =>
+ CryptoStreamUtils.createWritableChannel(out, conf, key)
+ }.getOrElse(out)
+ } catch {
+ case e: Exception =>
+ out.close()
+ throw e
+ }
+ }
+
+}
+
+private class EncryptedBlockData(
+ file: File,
+ blockSize: Long,
+ conf: SparkConf,
+ key: Array[Byte]) extends BlockData {
+
+ override def toInputStream(): InputStream =
Channels.newInputStream(open())
+
+ override def toManagedBuffer(): ManagedBuffer = new
EncryptedManagedBuffer()
+
+ override def toByteBuffer(allocator: Int => ByteBuffer):
ChunkedByteBuffer = {
+ val source = open()
+ try {
+ var remaining = blockSize
+ val chunks = new ListBuffer[ByteBuffer]()
+ while (remaining > 0) {
+ val chunkSize = math.min(remaining, Int.MaxValue)
+ val chunk = allocator(chunkSize.toInt)
+ remaining -= chunkSize
+
+ while (chunk.remaining() > 0) {
+ source.read(chunk)
--- End diff --
as mentioned above, needs EOF error handling.
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
