Skip to content

fix(deploy): L221 dedup_key on canary-rollback alert + lib pin v0.21→v0.24#287

Merged
cipher813 merged 2 commits into
mainfrom
feat/l221-canary-alert-dedup-key
May 22, 2026
Merged

fix(deploy): L221 dedup_key on canary-rollback alert + lib pin v0.21→v0.24#287
cipher813 merged 2 commits into
mainfrom
feat/l221-canary-alert-dedup-key

Conversation

@cipher813
Copy link
Copy Markdown
Owner

Summary

  • Adds `--dedup-key "canary-fail-${FUNCTION_NAME}-v${VERSION}"` to the canary-rollback alert publish in `infrastructure/deploy.sh`.
  • Lib pin v0.21.0 → v0.24.0 across requirements.txt + Dockerfile.
  • Closes L221 dedup retrofit slice for data repo (consumer migration Add PredictorHealthCheck to weekday pipeline #6, completes the 4-repo retrofit).

Why

The 2-day silent rollback chain that motivated L221 (10 consecutive rollbacks before operator notice) would have produced 10 alerts; now collapses to one alert per (Lambda, version) within the lib v0.24.0 60-min default window.

Originally filed as P1 owed BEFORE the 4 L221 canary PRs merged; those merged 5/21 23:42-23:51Z so this is a post-merge retrofit per the ROADMAP downgrade rule.

Test plan

  • Full suite — 1425 passed, 1 skipped
  • First production exercise: next failing canary on a data deploy

🤖 Generated with Claude Code

cipher813 and others added 2 commits May 22, 2026 06:24
…v0.24

Adds ``--dedup-key "canary-fail-${FUNCTION_NAME}-v${VERSION}"`` to the
canary-rollback alert publish in ``infrastructure/deploy.sh``. The
2-day silent rollback chain that motivated L221 (alpha-engine-data
#274 retrospective — 10 consecutive rollbacks across 2 days,
operator missed them) would have produced 10 alerts pre-fix; with
the lib v0.24.0 dedup substrate, an image-wide rebuild that breaks
N consecutive canaries within the hour collapses to one alert per
(Lambda, version).

Originally filed as a P1 owed BEFORE the 4 L221 canary PRs merged;
those PRs merged 5/21 23:42-23:51Z so this is a post-merge retrofit
per the ROADMAP's downgrade rule.

Lib pin v0.21.0 → v0.24.0 in lockstep across requirements.txt +
Dockerfile (the [flow_doctor]-only Lambda install line — the EC2
[arcticdb,flow_doctor,rag] install resolves transitively from
requirements.txt). v0.24.0 ships the ``--dedup-key`` CLI flag this
retrofit relies on.

Full suite: 1425 passed, 1 skipped.

Composes with [[reference_alpha_engine_lib_alerts_v0_24_0_dedup]]
(consumer migration #6 of the v0.24.0 substrate, post-merge slice
completing the 4-repo L221 retrofit).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
@cipher813 cipher813 merged commit b3c2b2b into main May 22, 2026
1 check passed
@cipher813 cipher813 deleted the feat/l221-canary-alert-dedup-key branch May 22, 2026 13:52
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant