Github user JoshRosen commented on a diff in the pull request:
https://github.com/apache/spark/pull/21346#discussion_r192566116
--- Diff:
common/network-common/src/main/java/org/apache/spark/network/client/TransportClient.java
---
@@ -141,26 +141,14 @@ public void fetchChunk(
StreamChunkId streamChunkId = new StreamChunkId(streamId, chunkIndex);
handler.addFetchRequest(streamChunkId, callback);
- channel.writeAndFlush(new
ChunkFetchRequest(streamChunkId)).addListener(future -> {
--- End diff --
Thanks for explaining. I guess the re-ordering of `channel.close()` and the
`handler` operations is safe because the handler doesn't hold references to the
channel / otherwise does not interact with it (and doesn't hold references to
objects tied to channel lifecycle (like buffers))?
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
