fix: recon protocol hang with large diffs #356
Merged
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
If two nodes each had 2K+ events that the other node did not have it was possible for the protocol to deadlock as both nodes tried to write those events without read the values from the other node. The fix was accomplished by splitting the protocol into two loops, a read loop and a write loop. They communicate with each other using message passing. The code now has both a clean separation of concerns as well as the behavior that we concurrently read and write from the network so we do not enter a state where we are exclusively writing to the network.
nathanielc
temporarily deployed
to
github-tests-2024
GitHub Actions
Inactive
nathanielc
temporarily deployed
to
github-tests-2024
GitHub Actions
Inactive
nathanielc
commented
May 17, 2024
| ) -> Result<()> { | ||
| // TODO: This logic has two potential failure modes we need to test them | ||
| // 1. We allocate memory of all keys in the range, this can be very large. | ||
| // 2. We spend a lot of time writing out to the stream but not reading from the stream. |
There was a problem hiding this comment.
Both of these points are addressed with this change.
nathanielc
temporarily deployed
to
github-tests-2024
GitHub Actions
Inactive
nathanielc
temporarily deployed
to
github-tests-2024
GitHub Actions
Inactive
dav1do
approved these changes
May 17, 2024
recon/src/protocol.rs
Outdated
| .await?; | ||
| for key in keys { | ||
| if let Some(value) = recon.value_for_key(key.clone()).await? { |
There was a problem hiding this comment.
If there isn't a value should we be concerned at all now that we should discover them together or am I forgetting about something?
There was a problem hiding this comment.
Good call, going to convert this into an error.
nathanielc
temporarily deployed
to
github-tests-2024
GitHub Actions
Inactive
nathanielc
temporarily deployed
to
github-tests-2024
GitHub Actions
Inactive
Merged
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.
If two nodes each had 2K+ events that the other node did not have it was possible for the protocol to deadlock as both nodes tried to write those events without read the values from the other node.
The fix was accomplished by splitting the protocol into two loops, a read loop and a write loop. They communicate with each other using message passing. The code now has both a clean separation of concerns as well as the behavior that we concurrently read and write from the network so we do not enter a state where we are exclusively writing to the network.