queso backport: Use randomly generated task IDs to avoid collisions when receiving messages from multiple workers #8710
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…ssages from multiple workers Previously, we were using a monotonically increasing ID that started out with 0. This means there was a high chance of duplicate task IDs being processed at the same time when different execution contexts (aka web workers) sent messages to the another context. Since we store tasks keyed by task ID, tasks that were sent later overwrote tasks with the same ID that were already waiting in the queue. In practice, this only affected messages sent from a worker to the main thread. In most browsers, processing these tasks on the main thread is very quick, so they don't spend a lot of time in the queue. While this bug affects all browsers, Internet Explorer in particular isn't the fastest when processing messages, so had the highest likelihood of message ID collisions actually overwriting tasks waiting in the queue. We didn't observe this bug while testing on CI because we fake the web worker environment there, and task IDs could never be duplicated because they all ran in the same execution context.
mourner
approved these changes
Aug 30, 2019
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Backports #8708 to
release-quesoPreviously, we were using a monotonically increasing ID that started out with 0. This means there was a high chance of duplicate task IDs being processed at the same time when different execution contexts (aka web workers) sent messages to the another context. Since we store tasks keyed by task ID, tasks that were sent later overwrote tasks with the same ID that were already waiting in the queue. In practice, this only affected messages sent from a worker to the main thread. In most browsers, processing these tasks on the main thread is very quick, so they don't spend a lot of time in the queue. While this bug affects all browsers, Internet Explorer in particular isn't the fastest when processing messages, so had the highest likelihood of message ID collisions actually overwriting tasks waiting in the queue. We didn't observe this bug while testing on CI because we fake the web worker environment there, and task IDs could never be duplicated because they all ran in the same execution context.
This fixes #8702
Needs backport to release-queso.