spark-instrumented-optimizer/common
Chandni Singh 6d88212f79 [SPARK-34840][SHUFFLE] Fixes cases of corruption in merged shuffle …
### What changes were proposed in this pull request?
This PR fixes bugs that causes corruption of push-merged blocks when a client terminates while pushing block. `RemoteBlockPushResolver` was introduced in #30062 (SPARK-32916).

There are 2 scenarios where the merged blocks get corrupted:
1. `StreamCallback.onFailure()` is called more than once. Initially we assumed that the onFailure callback will be called just once per stream. However, we observed that this is called twice when a client connection is reset. When the client connection is reset then there are 2 events that get triggered in this order.
 - `exceptionCaught`. This event is propagated to `StreamInterceptor`. `StreamInterceptor.exceptionCaught()` invokes `callback.onFailure(streamId, cause)`. This is the first time StreamCallback.onFailure() will be invoked.
 - `channelInactive`. Since the channel closes, the `channelInactive` event gets triggered which again is propagated to `StreamInterceptor`. `StreamInterceptor.channelInactive()` invokes `callback.onFailure(streamId, new ClosedChannelException())`. This is the second time  StreamCallback.onFailure() will be invoked.

2. The flag `isWriting` is set prematurely to true. This introduces an edge case where a stream that is trying to merge a duplicate block (created because of a speculative task) may interfere with an active stream if the duplicate stream fails.

Also adding additional changes that improve the code.

1.  Using positional writes all the time because this simplifies the code and with microbenchmarking haven't seen any performance impact.
2. Additional minor changes suggested by mridulm during an internal review.

### Why are the changes needed?
These are bug fixes and simplify the code.

### Does this PR introduce _any_ user-facing change?
No

### How was this patch tested?
Added unit tests. I have also tested these changes in Linkedin's internal fork on a cluster.

Co-authored-by: Chandni Singh chsinghlinkedin.com
Co-authored-by: Min Shen mshenlinkedin.com

Closes #31934 from otterc/SPARK-32916-followup.

Lead-authored-by: Chandni Singh <singh.chandni@gmail.com>
Co-authored-by: Min Shen <mshen@linkedin.com>
Signed-off-by: Mridul Muralidharan <mridul<at>gmail.com>
2021-03-25 12:47:46 -05:00
..
kvstore [SPARK-33662][BUILD] Setting version to 3.2.0-SNAPSHOT 2020-12-04 14:10:42 -08:00
network-common [SPARK-34722][CORE][SQL][TEST] Clean up deprecated API usage related to JUnit4 2021-03-14 23:33:03 -07:00
network-shuffle [SPARK-34840][SHUFFLE] Fixes cases of corruption in merged shuffle … 2021-03-25 12:47:46 -05:00
network-yarn [SPARK-33212][BUILD] Upgrade to Hadoop 3.2.2 and move to shaded clients for Hadoop 3.x profile 2021-01-15 14:06:50 -08:00
sketch [SPARK-33662][BUILD] Setting version to 3.2.0-SNAPSHOT 2020-12-04 14:10:42 -08:00
tags [SPARK-34578][SQL][TESTS][TEST-MAVEN] Refactor ORC encryption tests and ignore ORC shim loaded by old Hadoop library 2021-03-02 16:52:27 +09:00
unsafe [SPARK-34192][SQL] Move char padding to write side and remove length check on read side too 2021-01-26 02:08:35 +08:00