beacon: shard discover/relay/punch rate limiters (fix UDP recv-queue saturation) - #48
Merged
Conversation
…uration) Under a full-fleet reconverge flood the beacon's UDP recv-queue grew unbounded (60K->350K bytes, never draining) while the box had CPU headroom — the signature of lock contention. All 2*NumCPU reader goroutines serialized on three single global mutexes (discoverRateMu, relayRateMu, punchRateMu), each taken on every discover/relay/punch packet. Shard each rate limiter 256 ways (the endpoint store was already sharded), so readers touching different keys proceed in parallel. Deployed effect: recv-queue drains to 0 between bursts instead of growing; tunnel coordination keeps up under the flood. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Codecov Report✅ All modified and coverable lines are covered by tests. 📢 Thoughts on this report? Let us know! |
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Under a full-fleet reconverge flood the beacon's UDP recv-queue grew unbounded (60K→350K bytes, never draining) while the box had CPU headroom — classic lock contention. All 2×NumCPU reader goroutines serialized on three single global mutexes (
discoverRateMu,relayRateMu,punchRateMu), each taken on every discover/relay/punch packet. This shards each limiter 256 ways (the endpoint storenodeMapwas already sharded).Deployed effect (live): recv-queue now drains to 0 between bursts instead of growing without bound; the beacon keeps up with the reconverge flood so tunnel coordination proceeds. Full
-racesuite green.🤖 Generated with Claude Code