schedstat: add runnable goroutine spike detection#3
Merged
Conversation
tbg
force-pushed
the
runnable-threshold
branch
2 times, most recently
from
6758305 to
b4b7cba
Compare
Add detection of time windows where the runnable goroutine count exceeds a threshold (default 5*GOMAXPROCS, configurable via --runnable-threshold). Restructure the anomaly output into separate "Latency Spikes" and "Runnable Spikes" sections, each with per-spike root cause detail (burst breakdown, heavy unblockers, longest run, queue activity). Add experiment-upsert1000 gateway and leaseholder traces to the test corpus with golden file output. Code quality improvements from review: - Replace string-typed spikeType with const/iota enum - Fix fragile append (allSpikes could alias latencySpikes) - Pass top as parameter instead of reading global opts in query functions - Deduplicate count+data SQL queries using COUNT(*) OVER() - Assign spike display indices in analyze() instead of mutating in print - Log burst breakdown errors unconditionally instead of only under --verbose - Extract testOpts() helper to reduce test boilerplate - Use fixed goroutineThreshold in golden tests for determinism - Add comments on cumulative runnable count assumption and time semantics Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
tbg
force-pushed
the
runnable-threshold
branch
from
b4b7cba to
eb992fe
Compare
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Detect anomalies based on runnable goroutine count in addition to p99
latency spikes. The --runnable-threshold flag sets the count above which
a window is flagged (default: 5*GOMAXPROCS).
Latency spikes are listed first in the output, followed by runnable
count spikes marked with [gorunnable].
Also adds a deterministic sort to the heavy unblocker query to avoid
flaky golden test output.