Repository navigation
Conversation
…l-sync audit sink to error-helper
HouMinXi
marked this pull request as ready for review
November 1, 2026 09:13
This was referenced Nov 1, 2026
Contributor
Author
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
The model-sync timing wrapper added in #18466 hides the final catch block's
saveCallLogcall fromcheck:error-helper. The gate flags the internal audit entry even though the HTTP response still usessanitizeErrorMessage.The catch block now calls
saveCallLogdirectly insidetimings.measure("call-log", ...). The log fields, best-effort rejection handling and HTTP sanitizer stay unchanged. The timer now also includes construction of the log entry. No checker rules, allowlists, dependencies or environment settings change.Related: #17219 and the CI investigation in #18658. This patch does not address the separate environment-documentation failure.
Local validation
Based on
release/v3.8.53at139070680f77728e2309eb4df18463282a1c990a.tests/unit/model-sync-error-helper.test.ts: five tests cover the real route, timed internal logging and raw HTTP negative controls. The real-route test fails before the fix and passes after it.check-error-helper*.test.tsfiles pass with Node 24.22.1's native TypeScript support.Local checks used exact source files retrieved from GitHub, without a complete installed checkout. Project ESLint, Prettier, typecheck, hooks, the whole-repository gate and route integration tests were not run locally. Prettier and local hooks remain unverified.
CI after marking ready for review
These runs checked head
78ccd6387082f59f92fd819a7887d652d2769551against base139070680f77728e2309eb4df18463282a1c990a, using merge checkout4e8ea35bdf423cd7e25ff75a2bdb908a5373c7fd.Passing checks:
check:error-helperpassed across 1,405 files with the same 18 frozen exceptions. The containing Fast Quality Gates job failed on other checks listed below.sync-models-phase-timings.test.tstests passed, including real-route persistence, completed call-log timing, disabled instrumentation and the invalid-discovery 502 path.CI, Quality Gates and API Route Typecheck still fail. The failures below are reported in files or checks this two-file patch does not modify:
OMNIROUTE_MODEL_SYNC_TIMINGSis missing from.env.example; unit shards 3 and 7 fail the environment contract checks.file://${process.argv[1]}direct-run check inscripts/cli/omniroute-sso.mjs.ca7fc3f5ff3ec3e1f10e056faeb7332a568f6b28; five covering test entries are missing from mutation configuration; the vulnerability count is 29 against baseline 27; dashboard typecheck reportscliTools.tsTS2353. Provenance also stops the Lint job after ESLint; the vulnerability count also fails Extended Quality.availabilityRouterFilters.tsTS2345 andcliTools.tsTS2353 exceed the frozen baseline.provider-stats-error-state.test.tsx:43expects the alert to disappear after dismissal, but it remains.Unit shards 4, 5, 6 and 8 passed. Coverage was skipped after unit failures; aggregate CI and Quality admission failed. These runs do not establish that every remaining failure reproduces on the base alone, and the Vitest failure has not been classified as flaky. Merge readiness has not been verified.
Related baseline tracking: #17219.