Repository navigation
fix(benchmarks)!: preserve TCP kernel sample onset - #60
Merged
Merged
Conversation
Keep the first four qualifying snapshots and the latest four in each existing bounded history, preserving sampling and transport behavior. BREAKING CHANGE: opt-in tcp_kernel_sample_history records and the nested tcp_kernel_samples summary now declare schema 2 with first_4_latest_4 retention, sample ordinals and explicit policy metadata. Schema 1 raw histories remain tail-only evidence; consumers must interpret the recorded schema rather than relabel historical output.
4 tasks
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
TCP stall histories previously retained only their latest eight samples. A server that stayed alive after client cleanup could overwrite the observations needed to compare both peers before cancellation.
Keep the first four qualifying observations and the latest four in the existing bounded history. Record sample ordinals, omitted-middle counts and the existing local steady-clock origin for same-host correlation. Sampling cadence, eligibility, storage budget, transport, deadlines and workload remain unchanged. Update both documentation languages.
Validation:
This improves diagnostic evidence; it does not resolve or establish a cause for the 5,000-connection stall. Focused local checks do not replace the required hosted matrix. Raw evidence stays local.
BREAKING CHANGE: opt-in TCP kernel histories and their nested sampler summary use schema 2 with first_4_latest_4 retention and explicit steady-clock metadata. Diagnostic consumers must inspect the recorded schema. Existing schema 1 evidence remains tail-only and must not be relabeled. Library API and transport behavior are unchanged.