Files
awoooi/docs/security/MONITORING-OWNER-REQUEST-DRAFT.md
Your Name 8c1f9dca0f
All checks were successful
Code Review / ai-code-review (push) Successful in 13s
CD Pipeline / tests (push) Successful in 1m37s
CD Pipeline / build-and-deploy (push) Successful in 3m49s
CD Pipeline / post-deploy-checks (push) Successful in 1m29s
feat(iwooos): 強化告警鏈路 no-false-green gate
2026-06-15 17:45:28 +08:00

7.9 KiB
Raw Permalink Blame History

IwoooS Monitoring / Alerting / Observability owner request draft

項目 內容
日期 2026-06-14
狀態 owner_request_draft_ready_not_dispatched
工具 scripts/security/monitoring-owner-request-draft.py
Snapshot docs/security/monitoring-owner-request-draft.snapshot.json
Source inventory docs/security/monitoring-alerting-observability-inventory.snapshot.json
runtime gate 0

1. 目的

本文件把 Monitoring / Alerting / Observability repo-only 清冊中的 60 個 surface 轉成人工送件前 owner request draft讓 Prometheus、Alertmanager、Grafana、SigNoz、Sentry、Langfuse、OTEL、Telegram / notification policy、deploy / reload script 與 alert chain smoke script 都有一致的 owner 回覆欄位。

本階段仍是 metadata-only request draft不連 live Prometheus、不 reload Alertmanager、不套用 Grafana / SigNoz、不部署 Sentry、不改 Langfuse、不 reload OTEL、不改 receiver route、不建立 silence、不送 Telegram、不 fire live alert、不跑 alert chain smoke、不 SSH、不 kubectl、不讀 secret value、不寫 production。

2. 摘要

指標 目前值 說明
request draft 60 全部由 committed monitoring inventory surface 轉換
write-capable request draft 11 可能 reload、deploy、send notification、fire alert 或 restart exporter 的 surface
live evidence required request 60 每份 request 都需要 owner 提供脫敏 live evidence ref
request field 24 每份草稿固定欄位數
required owner field 14 owner 必須補齊的欄位數
blocked action 24 驗收前全部禁止
request sent / recipient confirmed 0 / 0 尚未送件,收件人也未確認
owner response received / accepted 0 / 0 不得假性拉高
live evidence received 0 未收到 live config hash、receiver diff 或 route smoke proof
reload / receiver / route smoke accepted 0 / 0 / 0 未授權 reload、receiver route change 或 live smoke
runtime gate / action button 0 / 0 不開任何執行入口

3. Request Draft 類型

類型 範圍 目前邊界
Prometheus config / rules base config、remote write、generated target、service registry、alert rules、drift guard 只收 owner ref不 reload、不 remote write change
Alertmanager receiver route、receiver、grouping source 只收 receiver owner / diff ref不改 receiver、不 reload
Grafana dashboard / rules dashboard JSON、alert rule 只收 dashboard owner / rollback ref不 import、不 apply
SigNoz / OTEL alert rule、log rule、API client、collector config 只收 pipeline owner / data export boundary不 apply、不 reload
Sentry / Langfuse compose、deploy、webhook receiver、API client、runbook 只收 live hash / upgrade window不 deploy、不 restart
Telegram / notification gateway service、failure-only policy、notification matrix、recurrence notifier 只收 receiver / receipt / noise budget owner不送訊息
deploy / reload scripts Alertmanager、Prometheus、Sentry、exporter deploy 或 reload-capable script 只建立 request draft不執行腳本
alert smoke live alert、test alert、alert chain smoke 只收 allowed receiver、execution window、cleanup owner不 fire alert

4. Owner 必填欄位

欄位 說明
owner_role_or_team 負責角色或團隊
decision owner 對此 surface 的判定
decision_reason 決策理由摘要,不得包含機敏值
affected_scope 受影響的 monitoring / notification / product scope
redacted_evidence_refs 脫敏 evidence id、文件路徑、ticket id 或 hash
live_config_hash_ref 脫敏 live config hash 或 owner-provided ref
reload_owner reload / deploy decision owner
receiver_owner receiver / route / receipt owner
route_smoke_plan route smoke 或 receipt proof 計畫
maintenance_window 維護窗口或禁止窗口
rollback_owner rollback 負責人與 rollback ref
validation_plan 驗證與 post-check plan
noise_budget_owner 告警噪音、silence、dedup 與測試通知負責人
followup_owner 後續補件或決策負責人

5. 禁止動作

prometheus_reload
alertmanager_reload
grafana_dashboard_apply
signoz_rule_apply
sentry_deploy
langfuse_config_change
otel_collector_reload
receiver_route_change
silence_policy_change
telegram_send
notification_route_change
webhook_receiver_change
remote_write_change
exporter_deploy
live_alert_fire
alert_chain_smoke
ssh_read
ssh_write
kubectl_action
secret_value_collection
host_write
active_scan
production_write
runtime_gate_open

6. 指令

固定 committed snapshot 時間:

python3 scripts/security/monitoring-owner-request-draft.py \
  --root . \
  --inventory-report docs/security/monitoring-alerting-observability-inventory.snapshot.json \
  --output docs/security/monitoring-owner-request-draft.snapshot.json \
  --generated-at 2026-06-14T23:25:00+08:00

只讀 guard

python3 scripts/security/security-mirror-progress-guard.py --root .
python3 scripts/security/source-control-owner-response-guard.py --root .

7. 完成度

工作 完成度 說明
owner request draft artifact 100% 60 份 request draft 已由 committed inventory 產生
request dispatch 0% 尚未送件recipient 未確認
owner response received / accepted 0% 尚未收到或接受任何 owner response
live evidence collection 0% 未讀 live monitoring stack
reload / receiver / route smoke acceptance 0% 未授權、未執行
runtime / production write 0% 無 action button無 production write

8. 邊界

這份 owner request draft 不是資安批准、不是 runtime approval、不是 alert chain smoke proof也不是 live monitoring truth。不得把 request draft、snapshot、IwoooS UI、AwoooP approval、LOGBOOK 或 committed source visibility 解讀成 Prometheus reload、Alertmanager reload、Grafana import、SigNoz apply、Sentry deploy、Langfuse change、OTEL reload、receiver route change、silence change、Telegram send、live alert fire、alert chain smoke、SSH、kubectl、active scan、secret collection、host write 或 production write 授權。

9. 2026-06-15 owner response acceptance 只讀帳本

2026-06-15 已新增 docs/security/MONITORING-OWNER-RESPONSE-ACCEPTANCE.mddocs/security/monitoring-owner-response-acceptance.snapshot.json,把本文件的 60 份 request draft 轉成 owner response acceptance 只讀帳本。固定 candidates=60write_capable=11live_evidence_required=60owner_fields=14reviewer_checks=15outcome_lanes=7blocked_actions=28owner response received / accepted / rejected、live evidence、reload、receiver route change、silence change、Telegram send、alert chain smoke、host write、runtime gate 與 action button 仍全部為 0 / false

2026-06-15 後續已在 acceptance ledger 補上 no-false-green 回補欄位,最新固定為 acceptance_fields=38reviewer_checks=23outcome_lanes=12blocked_actions=34;新增 incident context、alert chain health、receiver receipt、stale alert、silence / dedup、false-green risk、post-reload readback 與 cross-project notification ref。這仍不代表 owner response received / accepted、live alert chain truth、reload、Telegram send 或 runtime gate。

這份 acceptance ledger 只定義未來 owner response 如何收件、隔離、拒收、補件或進 monitoring reviewer review不代表 request sent、owner response received / accepted、Prometheus reload、Alertmanager reload、Grafana import、SigNoz apply、Sentry deploy、Langfuse change、OTEL reload、receiver route change、Telegram send、live alert fire、alert chain smoke 或 runtime gate。