Files
obsidian-vault/work/projects/cpm-web-extension-breakage-validation.md
T

6.4 KiB
Raw Blame History

CPM Web Extension Breakage — статус и PR-план

Актуально на 2026-09-22. Полная трассировка: cpm-web-extension-breakage-findings.
Experiment Definition: cpm-background-graveyard-experiment-definition.

Решение

Есть два независимых класса проблем:

  • Per-tab lifecycle исправлен в M1 и уже merged.
  • All-tabs CPM outage идёт stack: M3 telemetry → M4 hung recovery → M5 bounded graveyard → M6 A/B.

M4 перезагружает Web Extensions только после второго сигнала hung, когда монитор подтверждает global CPM outage. Первый сигнал остаётся диагностическим: он не меняет состояние расширений.

Статус проблем

Проблема Статус Что делаем
Forced background-process death → persistent all-tabs CPM outage Подтверждено M3 измеряет production incidence; M5 лечит; M6 сравнивает control/treatment.
Natural crash/memory/CPU termination → production outage Не проверено в production Один production release M3 до включения M5/M6.
Close Other Tabs, popup duplicate, pin/unpin/reorder/move/merge/window close Исправлено, merged M1 PR #6827.
Нестабильный Web Extension base URL Исправлено локально M2 задаёт deterministic base URL до load/reload и отвергает malformed identifier без crash.
Startup/session-restore content-script miss Подтверждено как transient Не объясняет persistent all-tabs outage.
Failed background load навсегда блокирует WebKit Отвергнуто WebKit повторяет load; нужна повторяемая причина/traffic.
Critical memory pressure как root cause Не воспроизведено Остаётся diagnostic correlation M3.
Event starvation / JSC hang / native messaging stall Не проверено Остаётся в M3 diagnostic buckets.
Network process restart Отвергнуто На проверенной сборке CPM восстановился.
removeAllUserScripts / flag transition / scriptlet reload / page JS как all-tabs cause Отвергнуто Это не persistent global failure mode.

Milestones

Milestone Branch / worktree Готовность
M1 tab lifecycle codex/cpm-tab-lifecycle-fixes Merged: PR #6827, CI green.
M2 stable base URL alex/cpm-stable-base-url — /Users/admin/DuckDuckGo/apple-browsers.git/cpm-stable-base-url Production-ready: 2 files; 11/11 targeted tests. Добавлена safe URL construction и negative validation вместо force-unwrap.
M3 attribution telemetry alex/cpm-failure-telemetry — /Users/admin/DuckDuckGo/apple-browsers.git/cpm-failure-telemetry Production-ready code: passive PII-free buckets, app-update/launch attribution, stable-base-URL transition fact, iOS/macOS schema validation green. Privacy triage: cpm-m3-telemetry-privacy-triage.
M4 hung recovery alex/fix-cpm-breakage — /Users/admin/DuckDuckGo/apple-browsers.git/cpm-breakage-pixel In progress: после второго сигнала hung перезагружает Web Extensions; первый сигнал только открывает диагностический episode.
M5 graveyard mitigation alex/cpm-graveyard-mitigation — /Users/admin/DuckDuckGo/apple-browsers.git/cpm-graveyard-mitigation Production-ready, default off: one retained view maximum, 3-second bounded hold, stale release cancellation, unload/reload/runtime kill-switch cleanup. 24/24 targeted tests; earlier real forced-kill iOS E2E recovered with release at ≈3.15 s.
M6 graveyard A/B alex/cpm-graveyard-ab — /Users/admin/DuckDuckGo/apple-browsers.git/cpm-graveyard-ab Production-ready code: lazy exposure enrollment, stable cohorts, control/treatment isolation, bounded pending queue, operational pixels and standard experiment_metrics outcome. 421/421 WebExtensions tests; iOS/macOS app builds; iOS/macOS pixel validators.

All milestone diffs pass git diff --check; changed Swift lines are within 150 columns.

Stack and Git state

Independent PRs:

  1. M2 — stable base URL.
  2. M3 — production telemetry.
  3. M4 — reload after confirmed hung (the second signal).

Stacked PRs:

M3 telemetry
└── M4 hung recovery
    └── M5 graveyard mitigation
        └── M6 A/B experiment

Current refs intentionally preserve the earlier no-commit workflow:

  • M3 and M5 branch refs: 814bce5d11; their milestone changes are unstaged.
  • M5 ghost commit used only as M6 base: cefad244c5.
  • M6 branch ref: cefad244c5; M6 changes are unstaged above it.
  • M5’s final combined pixel description has been pulled into M6.
  • Local origin/main advanced from the M3 merge base by one unrelated pixel-definition commit (67a2f2b7b2); it does not overlap CPM files.

Before opening PRs: commit M2 and M3, update them onto current origin/main, recreate/rebase the M5 commit onto M3, then rebase M6 onto that M5 commit. Do not include generated Package.resolved churn; Cursor currently regenerates it in the open M3 workspace.

Experiment contract

  • Experiment: cpmBackgroundGraveyardExperiment.
  • Cohorts: control, treatment.
  • Enrollment: only on an actual eligible CPM background-process termination.
  • Primary framework metric: cpmBackgroundGraveyardOutcome, window 0...1, values healthy, initialization_failed, not_measured.
  • Rollback: disable both cpmBackgroundGraveyardExperiment and cpmBackgroundGraveyard.

Full definition and config draft: cpm-background-graveyard-experiment-definition.

Remaining gates

  1. Build and validate M4, then open M2/M3/M4/M5/M6 PRs in the topology above.
  2. Create separate iOS and macOS experiment-log tasks.
  3. Create Privacy Configuration + privacy/schema review tasks.
  4. Create Data Science sample-size/design and Native Apps Experiment Analysis task.
  5. Ship M3 first; enable M5/M6 only if production telemetry confirms the process-death hypothesis.

No Asana tasks, comments, or configuration were created during this review.