fix(observability): document real outcome vocabulary, add graphrunner lifecycle metrics (audit #11) - #259
Open
sbalabanov wants to merge 2 commits into
Open
fix(observability): document real outcome vocabulary, add graphrunner lifecycle metrics (audit #11)#259sbalabanov wants to merge 2 commits into
sbalabanov wants to merge 2 commits into
Conversation
|
|
xytan0056
approved these changes
Jul 31, 2026
sbalabanov
commented
Jul 31, 2026
… lifecycle metrics (audit #11) The metrics doc described outcome values (success/failure/cancelled) that did not match the code, which emits success/cancelled/user/infra/infra_retryable via ErrorCode.String(). "failure" was never emitted. The doc also showed a fictitious cache_read sub-op example and omitted the failures{error_code} counter and three cache-lookup counters. - Rewrite the outcome-vocabulary section with the real values and their derivation from ErrorCode.String() - Document the controller failures{error_code} counter - Document the three cache-lookup counters (treehash, graph, compared_targets) with hit/miss/silent-on-infra semantics - Replace the fictitious cache_read example with the real cache-lookup pattern - Fix stale CamelCase op-name comment in names.go (real ops are snake_case) - Add Begin/Complete lifecycle metrics to nativeGraphRunner.Compute with repo-tagged emitter - Add RecordCacheLookup graph_cache_lookup to the controller graph download path with correct hit/miss/silent-infra semantics - Add tests for graphrunner lifecycle metrics (success + error) and controller graph_cache_lookup counters (hit + miss) Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
sbalabanov
force-pushed
the
audit/fix-11-metrics-doc-and-lifecycle
branch
from
July 31, 2026 21:17
aeefa21 to
7721e00
Compare
sbalabanov
marked this pull request as ready for review
July 31, 2026 21:19
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
4 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.
Summary
Align observability documentation and instrumentation with Tango's actual result vocabulary and shared lifecycle metric pattern.
Intent
Ensure metrics consumers see the real
resultvalues (success,cancelled,user,infra, andinfra_retryable), add missing graph-runner lifecycle and graph-cache lookup instrumentation, and keep controller failures on the same standard finish histogram used by other components.Changes
ErrorCode.String().RecordCacheLookuphit/miss pattern without cataloging current call sites.Begin/Completelifecycle metrics tonativeGraphRunner.Computewith a repo-tagged scope.failures{error_code}counter in favor of the standardfinish{result=...}histogram.observability/metrics/names.go.Test Plan
go test ./controller/...go test ./graphrunner/...go test ./observability/metrics/... ./orchestrator/...make gazelleaifx validate --base origin/main --timeout 15 --overall-timeout 20 --no-cacheRevert Plan
Revert the commits from this PR to restore the prior documentation and instrumentation behavior.
Jira Issues
None.