docs(awooop): record t151 execution evidence [skip ci]

This commit is contained in:
Your Name
2026-05-24 14:55:05 +08:00
parent cd81d604d9
commit 4874f2b649
2 changed files with 96 additions and 0 deletions

View File

@@ -2665,6 +2665,13 @@ Phase 6 完成後
- 判讀T135 已把 runner ownership 從雙 runner 搶工收斂到 host runner 單一主控;下一段不要重新啟用 Docker-wrapped runner而是做 runner pool / repo label 隔離、API image `apt-get` / `chown -R` 分層、Web build cache/offload、Playwright apt source-list hygiene。
- 目前進度更新AwoooP 告警可觀測鏈約 99.998%Incident-level source correlation 可見性約 98.8%Source correlation apply 狀態鏈可驗證性約 99.72%Source correlation freshness / rolling gate 約 98.2%;前端 AI 自動化管理介面同步約 99.999%Dashboard snapshot / SSE console noise 收斂約 99.2%CI/CD runner hygiene 約 99.2%Runner ownership 收斂約 96%Build host pressure治理約 82%;完整 AI 自動化管理產品化約 99.960%。
**T151 Execution backend / Ansible evidence surfaced2026-05-24 台北)**
- 觸發使用者持續追問「Ansible 是否有被完整發揮」、「AI 自動化到底跑到哪個階段」、「前端是否同步呈現」。T147-T150 已讓首頁證據卡不被慢 endpoint 拖垮、模型路由可見、rollout-risk 可列 ArgoCD resource但首頁仍看不到 execution backend / Ansible attribution 的總覽。
- 修正:`truth-chain/quality/summary` 新增 `execution_backend_summary`,彙總 `operation_records_total``effective_execution_records_total``audit_only_operation_records_total``auto_repair_execution_records_total``ansible_considered_total``ansible_audit_record_total``ansible_candidate_total``ansible_check_mode_total``ansible_apply_total``ansible_pending_check_mode_total`。首頁 `AutomationEvidenceCard` 同步顯示執行後端列。此為 read-only truth-chain 彙總,沒有啟用 Ansible apply。
- Verificationlocal focused pytest 25 passedruff / py_compile / i18n JSON parse / `git diff --check` passlocal web typecheck 因 temp worktree 無 `node_modules/tsc` 跳過,由 Gitea Docker build 覆蓋。`dc09dac4 feat(awooop): surface execution backend evidence` 已推 Gitea main#2917 tests/build/post-deploy success、#2918 code-review successdeploy marker `cd81d604`。Production API/Web image 均為 `dc09dac4...` readyWorker readyArgoCD `sync=Synced health=Healthy revision=cd81d604...`。Production quality summary 回 `operation_records_total=3``effective_execution_records_total=0``audit_only_operation_records_total=3``auto_repair_execution_records_total=2``ansible_considered_total=3``ansible_audit_record_total=3``ansible_candidate_total=30``ansible_check_mode_total=0``ansible_apply_total=0``ansible_pending_check_mode_total=3`。Browser 驗證首頁顯示「執行證據:操作 3有效 0 / 稽核 3自動修復 2Ansible 稽核 3候選 30check-mode 0apply 0待接線 3」console errors 0。
- 判讀:現在前端可以清楚看到 Ansible 有被列為候選並寫入稽核,但尚未真正跑 check-mode / apply這解釋為什麼目前不能宣稱完整自動修復。下一段若要提高 Execution evidence / Ansible attribution應接 `approval_execution -> Ansible check-mode -> automation_operation_log -> verifier` 的乾跑鏈,不直接做 apply。Production health 仍 degraded原因是 GCP-A offlineAI route 正確落到 GCP-B111/Gemini 維持 fallback。
- 目前進度更新AwoooP 告警可觀測鏈約 99.998%;前端 AI 自動化管理介面同步約 99.9998%;首頁證據卡可用性約 99.7%AI route status 響應保護約 96%AI provider runtime health 約 60%GCP-A downGCP-B/111 可用Pipeline / 部署階段可觀測性約 95.7%Deploy rollout-risk 可觀測性約 99.1%Execution evidence / Ansible attribution 約 68%;完整 AI 自動化管理產品化約 99.984%。
**T150 CD rollout-risk resource evidence2026-05-24 台北)**
- 觸發T149 已清掉舊 CronJob failed history 噪音ArgoCD application 回到 Healthy但 CD rollout-risk summary 仍只有 top-level `argocd_health_not_healthy health=Degraded`,下一次 Degraded 時 Telegram / AwoooP 仍看不到是哪個 ArgoCD resource 節點不健康。原通知標題 `AWOOOI 部署風險已恢復` 也容易誤導,實際語意應是「部署完成但仍有風險證據」。
- 修正:`.gitea/workflows/cd.yaml` 新增 `collect_argocd_resource_evidence()`,在 ArgoCD 已同步但 health 非 Healthy 時,從 Application `.status.resources` 讀取前 5 個 `sync != Synced``health != Healthy` 節點,寫入 `resources=Kind/name ns=... sync=... health=... msg=...`;若 top-level Degraded 但沒有可見節點,寫 `resources=none_visible`。rollout-risk 通知標題改為 `AWOOOI 部署完成但仍有風險證據`