diff --git a/docs/LOGBOOK.md b/docs/LOGBOOK.md index 506ffc5bd..ada899e0a 100644 --- a/docs/LOGBOOK.md +++ b/docs/LOGBOOK.md @@ -54266,9 +54266,9 @@ production browser smoke: **完成內容**: - 新增 `scripts/reboot-recovery/windows99-vmx-source-locate-check.ps1` 與 `collect-windows99-vmx-source-locate-check.sh`,把 111 VMX source 定位從一次性手動查找提升為 no-write / no-secret / no-power-change check-mode collector。 - collector 只使用 BatchMode public-key SSH,可經 `--via-host wooo@192.168.0.110` 進入 Windows99,並以 stdin PowerShell bootstrap 執行 bounded path / VMX identity scan;不讀密碼、不寫遠端檔、不啟動或關閉 VM、不重啟 Windows service / host、不套 registry。 -- 2026-07-03 12:00 live production readback:SLO scorecard `active_blocker_count=8`、readiness `60%`、primary `fresh_all_host_reboot_event_missing`;priority readback 顯示 Windows99 VMX repair stage `source_repair_check_mode_ready`、next executable step `locate_or_relink_missing_vmx_source_check_mode`。 +- 2026-07-03 12:34 deploy 後 production readback:SLO scorecard `active_blocker_count=10`、readiness `60%`、primary `reboot_event_required_host_unreachable`;priority readback 顯示 Windows99 VMX repair stage `source_repair_check_mode_ready`、next executable step `locate_or_relink_missing_vmx_source_check_mode`,locator endpoint 顯示 `status=check_mode_locator_ready_windows99_vmx_source_required`、`missing_vmx_aliases=["111"]`、safe command `bash scripts/reboot-recovery/locate-windows99-vmx-source.sh --collect`。 - live locator 結果:expected 111 VMX path `exists=0`、`candidate_vmx_count=6`、`alias_hint_candidate_count=0`、`identity_target_hint_count=0`、`identity_unassigned_ubuntu_candidate_count=1`、`locate_status=single_identity_unassigned_ubuntu_candidate_requires_confirmation`。唯一 generic Ubuntu candidate 尚不能安全 relink;需要 external verified backup artifact 或更強 identity evidence 後才可產生 scoped relink / restore dry-run 與 post-verifier。 -- `FULL-STACK-COLD-START-SOP`、`REBOOT-RECOVERY-SOP` 與統帥插入需求台帳已更新,將 stale 10-blocker 狀態覆蓋為目前 8-blocker truth,並把下一步固定為 111 VMX source verified artifact / identity evidence。 +- `FULL-STACK-COLD-START-SOP`、`REBOOT-RECOVERY-SOP` 與統帥插入需求台帳已更新,將 deploy 前 8-blocker 讀回修正為 deploy 後 production 10-blocker truth,並把下一步固定為 111 VMX source verified artifact / identity evidence。 **本地驗證結果**: - `bash -n scripts/reboot-recovery/collect-windows99-vmx-source-locate-check.sh`:通過。 diff --git a/docs/runbooks/FULL-STACK-COLD-START-SOP.md b/docs/runbooks/FULL-STACK-COLD-START-SOP.md index eeb44fd28..92177809e 100644 --- a/docs/runbooks/FULL-STACK-COLD-START-SOP.md +++ b/docs/runbooks/FULL-STACK-COLD-START-SOP.md @@ -60,7 +60,7 @@ v1.113 Windows99 VMX source repair package API rule:production `/api/v1/agents v1.114 Windows99 VMX package public redaction boundary rule:public `/api/v1/agents/reboot-auto-recovery-slo-scorecard` 與 `/api/v1/agents/windows99-vmx-source-repair-package` 會先套用 LAN topology redaction,因此 VMX path 內的 `192.168.0.111` 可能在 public JSON 變成 `host:dev-a` 類 display alias。public response 必須同步輸出 `public_lan_topology_redacted=true`、`redacted_public_display_only=true`、`check_mode_probe_commands_executable=false` 與 `internal_console_package_required_for_execution=true`;所有 `vm_rows[]` 也必須帶相同 executable=false 邊界。任何 `host:...` VMX path 只能供瀏覽器 / Telegram / UI 顯示,不得作為 Windows console 可執行命令。要跑 `Test-Path` 時只能使用授權內部 console / no-secret package 保留的原始 Windows path;仍不得讀密碼、啟動 VM、重啟 service / host 或套用 registry。 -v1.115 Windows99 VMX source locate check-mode rule:2026-07-03 12:00 production readback 已更新為 `status=blocked_reboot_auto_recovery_slo_not_ready`、`active_blocker_count=8`、`readiness_percent=60`、`primary_blocker=fresh_all_host_reboot_event_missing`、`can_claim_all_services_recovered_within_target=false`,priority readback 同步顯示 `windows99_vmx_source_repair_action_stage=source_repair_check_mode_ready` 與 `windows99_vmx_source_repair_next_executable_step=locate_or_relink_missing_vmx_source_check_mode`。定位 111 VMX source 時固定使用 `scripts/reboot-recovery/collect-windows99-vmx-source-locate-check.sh --via-host wooo@192.168.0.110 --include-identity-metadata`;此 collector 只用 BatchMode public-key SSH、stdin PowerShell bootstrap 與 bounded VMX identity keys,不讀 secret、不寫遠端檔、不啟動 / 關閉 VM、不重啟 Windows service / host、不套 registry。live 結果:expected 111 VMX path `exists=0`、`candidate_vmx_count=6`、`alias_hint_candidate_count=0`、`identity_target_hint_count=0`、`identity_unassigned_ubuntu_candidate_count=1`、`locate_status=single_identity_unassigned_ubuntu_candidate_requires_confirmation`。該 generic Ubuntu candidate 不能直接 relink;先前 MAC / neighbor evidence 未證明它是 111,且 99 local named backup roots 與 110 bounded backup roots 未找到 111 VMX artifact。下一步只能是取得外部 verified backup artifact 或更強身份證據後建立 scoped relink / restore dry-run,再跑 post-verifier;不得用猜測候選直接 relink、power on 或宣稱 SLA 完成。 +v1.115 Windows99 VMX source locate check-mode rule:2026-07-03 12:34 deploy 後 production readback 已更新為 `status=blocked_reboot_auto_recovery_slo_not_ready`、`active_blocker_count=10`、`readiness_percent=60`、`primary_blocker=reboot_event_required_host_unreachable`、`can_claim_all_services_recovered_within_target=false`;priority readback 同步顯示 `windows99_vmx_source_repair_action_stage=source_repair_check_mode_ready`、`windows99_vmx_source_repair_next_executable_step=locate_or_relink_missing_vmx_source_check_mode`、`windows99_vmx_source_locator_status=check_mode_locator_ready_windows99_vmx_source_required`,locator endpoint 顯示 `missing_vmx_aliases=["111"]` 與 safe command `bash scripts/reboot-recovery/locate-windows99-vmx-source.sh --collect`。定位 111 VMX source 時固定使用 no-secret locator / collector;本輪 live check-mode 結果:expected 111 VMX path `exists=0`、`candidate_vmx_count=6`、`alias_hint_candidate_count=0`、`identity_target_hint_count=0`、`identity_unassigned_ubuntu_candidate_count=1`、`locate_status=single_identity_unassigned_ubuntu_candidate_requires_confirmation`。該 generic Ubuntu candidate 不能直接 relink;先前 MAC / neighbor evidence 未證明它是 111,且 99 local named backup roots 與 110 bounded backup roots 未找到 111 VMX artifact。下一步只能是取得外部 verified backup artifact 或更強身份證據後建立 scoped relink / restore dry-run,再跑 post-verifier;不得用猜測候選直接 relink、power on 或宣稱 SLA 完成。 2026-07-02 110 control-path / Harbor recovery receipt rule:若 Gitea Harbor repair queue 仍保留 `harbor_110_remote_ssh_publickey_auth_stalled`、remote-control unavailable、jobs stale 或 historical failure,但同一輪本地證據同時證明 `wooo` command path ready、110 local Harbor `/v2/` ready、public/internal registry `/v2/` 回 `401`,則該 Gitea Harbor repair 失敗只能列為 historical queue metadata,不得再當成 current SSH blocker。必須用 `/api/v1/agents/harbor-registry-controlled-recovery-receipt` 或同等 validator 合併 `diagnose-110-ssh-publickey-auth.sh`、`recover-110-control-path-and-harbor-local.sh --check`、public Gitea queue readback 與 registry `/v2/` verifier,並把機器可讀結果寫入 `docs/operations/harbor-110-control-path-recovery-readback-2026-07-02.snapshot.json` 類型的 snapshot。2026-07-02 live receipt 顯示:public/internal registry `/v2/` 均為 `401`、latest visible CD `#4335` 為 `Success`、Gitea Harbor repair failure 已是 `historical_after_latest_cd_success=true`;active blockers 收斂為 110 controlled CD lane config / binary / registration / service guardrail、active action container pressure,以及 Gitea CD jobs head-SHA / stale readback mismatch。若 local-console output 只有 `AWOOOI_110_CONTROLLED_CD_LANE_READY` marker,non110 runner parser 不得從 110 `BLOCKER` 行推導 non110 blocker;non110 只有看到 `AWOOOI_NON110_RUNNER_READY` marker 才能列入 active blocker。 diff --git a/docs/runbooks/REBOOT-RECOVERY-SOP.md b/docs/runbooks/REBOOT-RECOVERY-SOP.md index e246380a1..37595e2c7 100644 --- a/docs/runbooks/REBOOT-RECOVERY-SOP.md +++ b/docs/runbooks/REBOOT-RECOVERY-SOP.md @@ -3,7 +3,7 @@ > **版本**: v5.8 > **最後更新**: 2026-07-03 (台北時間) > **更新者**: Codex -> **觸發事件**: 2026-07-03 12:00 production SLO scorecard + Windows99 VMX source locate check-mode:111 VMX source 尚未找到,10 分鐘 SLO 仍 blocked +> **觸發事件**: 2026-07-03 12:34 deploy 後 production SLO scorecard + Windows99 VMX source locator readback:111 VMX source 尚未找到,10 分鐘 SLO 仍 blocked --- @@ -21,9 +21,9 @@ ## 架構概覽與依賴圖 -### 2026-07-03 12:00 P0 現況覆蓋 +### 2026-07-03 12:34 P0 現況覆蓋 -目前不得宣稱「全主機重啟後 10 分鐘內全部恢復」。production SLO scorecard 仍為 `blocked_reboot_auto_recovery_slo_not_ready`,`active_blocker_count=8`、`readiness_percent=60`、primary blocker `fresh_all_host_reboot_event_missing`,`can_claim_all_services_recovered_within_target=false`。8 個 active blockers 固定為: +目前不得宣稱「全主機重啟後 10 分鐘內全部恢復」。deploy marker `4e7f10cfd chore(cd): deploy ed419da [skip ci]` 後,production SLO scorecard 仍為 `blocked_reboot_auto_recovery_slo_not_ready`,`active_blocker_count=10`、`readiness_percent=60`、primary blocker `reboot_event_required_host_unreachable`,`can_claim_all_services_recovered_within_target=false`。10 個 active blockers 固定為: - `all_required_hosts_not_in_10_minute_reboot_window` - `fresh_all_host_reboot_event_missing` @@ -33,18 +33,20 @@ - `reboot_event_required_host_unreachable` - `windows99_vmware_guest_power_not_ready` - `windows99_vmware_vmx_missing` +- `windows99_vmware_autostart_config_not_ready` +- `windows99_update_no_auto_reboot_policy_not_ready` 最新 no-write artifact 以 production API readback、110 上最新 Windows99 collector readback 與本輪 VMX locate collector 為來源: -- 99 no-secret VMware verify stdout 已能透過 110 / public-key SSH 收集;`VMAuthdService`、`VMnetDHCP`、`AWOOOI-Start-VMware-VMs` 與 Windows Update no-auto-reboot policy 讀回為 ok。 +- 99 no-secret VMware verify stdout 已能透過 110 / public-key SSH 收集,但 deploy 後 production scorecard 仍上卷 `windows99_vmware_autostart_config_not_ready` 與 `windows99_update_no_auto_reboot_policy_not_ready`;未清除前不得把服務 / policy 視為綠燈。 - 111 VMX expected path 不存在;`windows99_vmware_vmx_missing` 仍是 Windows99 主要 source blocker。 - 111 / 188 / 120 / 121 / 112 VM power readback 仍為 not running;`windows99_vmware_guest_power_not_ready` 仍 blocked,且此 SOP 不授權直接 power change。 - 111 host 仍 unreachable。 - 112 / 120 / 121 可達,但 uptime 已超過 10 分鐘目標窗口。 - 188 可達,但 `systemd_state=degraded` 且 `awoooi-startup.service failed`。 -- `fresh_all_host_reboot_event_missing` 仍為 primary blocker,因此 10 分鐘 SLO 尚未證明。 +- `reboot_event_required_host_unreachable` 仍為 primary blocker,因此 10 分鐘 SLO 尚未證明。 -固定下一步:`locate_or_relink_missing_vmx_source_check_mode`。執行 `scripts/reboot-recovery/collect-windows99-vmx-source-locate-check.sh --via-host wooo@192.168.0.110 --include-identity-metadata`;本輪 live 結果為 expected 111 VMX path `exists=0`、`candidate_vmx_count=6`、`alias_hint_candidate_count=0`、`identity_target_hint_count=0`、`identity_unassigned_ubuntu_candidate_count=1`、`locate_status=single_identity_unassigned_ubuntu_candidate_requires_confirmation`。唯一 generic Ubuntu candidate 還不能安全 relink,因為 display name / alias 沒有 111 identity hint,先前 MAC / neighbor evidence 也未證明它是 111,且 99 local named backup roots 與 110 bounded backup roots 尚未找到 111 VMX artifact。下一步只能取得外部 verified backup artifact 或更強 identity evidence,再產生 scoped relink / restore dry-run 與 post-verifier。若 public API response 顯示 `public_lan_topology_redacted=true` 或 `check_mode_probe_commands_executable=false`,其中 `host:...` VMX path 只能當 public display evidence,不得直接貼到 Windows console 執行;要執行 probe 時必須使用授權內部 console / no-secret package 保留的原始 Windows path。禁止讀 Windows 密碼、啟動 / 關閉 VM、host reboot、service restart、Docker / Nginx / K3s / DB / firewall restart、restore、prune、delete。 +固定下一步:`locate_or_relink_missing_vmx_source_check_mode`。production `/api/v1/agents/windows99-vmx-source-locator-readback` 已回 `status=check_mode_locator_ready_windows99_vmx_source_required`、`missing_vmx_aliases=["111"]`、`safe_next_step=bash scripts/reboot-recovery/locate-windows99-vmx-source.sh --collect`、`secret_value_read=false`、`remote_write_performed=false`、`vm_power_change_allowed=false`。本輪 live check-mode 結果為 expected 111 VMX path `exists=0`、`candidate_vmx_count=6`、`alias_hint_candidate_count=0`、`identity_target_hint_count=0`、`identity_unassigned_ubuntu_candidate_count=1`、`locate_status=single_identity_unassigned_ubuntu_candidate_requires_confirmation`。唯一 generic Ubuntu candidate 還不能安全 relink,因為 display name / alias 沒有 111 identity hint,先前 MAC / neighbor evidence 也未證明它是 111,且 99 local named backup roots 與 110 bounded backup roots 尚未找到 111 VMX artifact。下一步只能取得外部 verified backup artifact 或更強 identity evidence,再產生 scoped relink / restore dry-run 與 post-verifier。若 public API response 顯示 `public_lan_topology_redacted=true` 或 `check_mode_probe_commands_executable=false`,其中 `host:...` VMX path 只能當 public display evidence,不得直接貼到 Windows console 執行;要執行 probe 時必須使用授權內部 console / no-secret package 保留的原始 Windows path。禁止讀 Windows 密碼、啟動 / 關閉 VM、host reboot、service restart、Docker / Nginx / K3s / DB / firewall restart、restore、prune、delete。 ### 五主機全貌 diff --git a/docs/workplans/2026-07-02-commander-inserted-requirements-priority-ledger.md b/docs/workplans/2026-07-02-commander-inserted-requirements-priority-ledger.md index b24a57cb6..1895db5cf 100644 --- a/docs/workplans/2026-07-02-commander-inserted-requirements-priority-ledger.md +++ b/docs/workplans/2026-07-02-commander-inserted-requirements-priority-ledger.md @@ -57,14 +57,14 @@ | 順序 | ID | 優先序 | 使用者插入要求 | 正規化工作項 | 目前狀態 | 下一個可驗證動作 | | --- | --- | --- | --- | --- | --- | --- | -| 1 | CIR-P0-RBT-001 | P0 | 「主機重啟後 10 分鐘內全部恢復,且要自動判斷所有主機被重啟」 | 建立 99/110/111/112/120/121/188 reboot event detector + 10 分鐘 SLO scorecard + fixed triage order | 2026-07-03 12:00 production scorecard:readiness `60%`、active blockers `8`、primary `fresh_all_host_reboot_event_missing`、`can_claim_all_services_recovered_within_target=false`;Windows99 no-secret collector 已收 verify stdout,Windows Update / VMware services / scheduled task 讀回 ok,但 111 VMX source missing 與 guest power not ready 仍 blocked | 優先收斂 111 VMX source locate / verified backup artifact、111 reachability 與 fresh all-host reboot event;不可宣稱 10 分鐘 SLA 已證明,不得用 RDP clipboard 片段或 generic VMX guess 當完成證據 | +| 1 | CIR-P0-RBT-001 | P0 | 「主機重啟後 10 分鐘內全部恢復,且要自動判斷所有主機被重啟」 | 建立 99/110/111/112/120/121/188 reboot event detector + 10 分鐘 SLO scorecard + fixed triage order | 2026-07-03 12:34 deploy 後 production scorecard:readiness `60%`、active blockers `10`、primary `reboot_event_required_host_unreachable`、`can_claim_all_services_recovered_within_target=false`;locator endpoint 已 ready,missing alias `111`,但 VMX source / guest power / autostart config / Windows Update policy / reboot event 仍 blocked | 優先收斂 111 VMX source locate / verified backup artifact、111 reachability 與 fresh all-host reboot event;不可宣稱 10 分鐘 SLA 已證明,不得用 RDP clipboard 片段或 generic VMX guess 當完成證據 | | 2 | CIR-P0-RBT-002 | P0 | 「沒有偵測到主機重啟」 | 修正 host reboot/shutdown/up detection:boot_id / uptime / node exporter / Windows exporter / VMware VM power state 都要進同一事件 | Scorecard 已接 collection packet + management probe;08:23 host probe 有 7 列但 99 只有 ping/RDP reachable、uptime unknown,111 unreachable,112/120/121/188 uptime 都已超過 10 分鐘窗口 | 讓 99 verifier / Windows exporter 或等效 no-secret readback 進入 host boot event,並補 111 reachability 證據 | | 3 | CIR-P0-RBT-003 | P0 | 「192.168.0.99 VMWare 要自動啟動,裡面 111/188/120/121/112 也自動啟動」 | Windows 99 VMware host autostart + guest VM autostart contract;VM host 111/188/120/121/112 開機順序與 readback | Source verifier / parser / API readback / collection packet 已完成;12:00 API / priority readback 顯示 `source_repair_check_mode_ready`、下一步 `locate_or_relink_missing_vmx_source_check_mode`;本輪 no-write locator 確認 expected 111 VMX path `exists=0`、`candidate_vmx_count=6`、`alias_hint_candidate_count=0`、`identity_target_hint_count=0`、`identity_unassigned_ubuntu_candidate_count=1`、`locate_status=single_identity_unassigned_ubuntu_candidate_requires_confirmation`;VMware services / scheduled task / Windows Update policy 已 ok,但 guest power 仍 not ready | 不得直接 relink generic Ubuntu candidate;先取得 external verified backup artifact 或更強 identity evidence,再做 scoped relink / restore dry-run 與 post-verifier;不得從 public redacted path 或 scorecard 直接啟動 / 關閉 VM、重啟 Windows service、讀密碼或 host reboot | | 4 | CIR-P0-RBT-004 | P0 | 「192.168.0.99 不可因 Windows Update 無預警重開」 | Windows Update reboot policy:active hours / no auto-restart / maintenance window / update notification audit | 10:15 source/API patch:若 nested verifier 回 `policy_blockers=["windows_update_policy_readback_missing"]`,API 必須 fail-closed 重新提升 `windows99_update_no_auto_reboot_policy_not_ready`;不得只信舊 Prometheus `windows99_update_no_auto_reboot_ready=true` | 補 no-secret policy readback / drift alert;若 policy 不綠,另走 controlled apply package + rollback + post-verifier,禁止要求或記錄 Windows 密碼,禁止由 scorecard 直接 apply registry | | 5 | CIR-P0-RBT-005 | P0 | 「網站重啟後 502 嚴重影響體驗,要維護頁,外部雲端或專業做法」 | Public maintenance fallback:Nginx / edge / external static maintenance page / status page / fail-open UX,避免 502 直出 | Source + runtime verifier 已實作;Gitea CD `#4459` Success、deploy marker `8d7a6faaf`,production API scorecard 已讀回 `public_maintenance_fallback.ready=true`、raw 5xx=`0`、unreachable without L1=`0`,P0 blockers `11`、readiness `47`;仍不可宣稱整體 10 分鐘 SLO 完成 | 若未來 public route 出現 raw 5xx,先走 L0 Nginx intercept / header verifier;若 edge unreachable,才產生 L1 external static origin / CDN decision record + rollback | | 6 | CIR-P0-RBT-006 | P0 | 「所有主機關機立刻 Telegram 告警,重啟後也要告警,其他告警一併完整思考」 | Down / shutdown suspected / reboot detected / reboot recovered / SLO missed / backup failed / freshness stale / CPU pressure / Gitea queue 告警矩陣 | HostDown / HostRebootEventDetected / RebootAutoRecoverySLOMissed 已存在;per-blocker reboot alerts 與 backup receipt rules 已 deploy/readback。Backup receipt 缺段已從 100 條 stage 噪音收斂為 110 / 188 兩條 host-level pending;仍需完整 shutdown/up E2E receipt | 補 Prometheus / Alertmanager active/resolved 與 outbound receipt;backup alert 先補 `/backup/alert-receipts/*.last_success` 脫敏 marker,不送測試 secret、不重啟主機 | | 7 | CIR-P0-RBT-007 | P0 | 「所有備份包含主機、DB、網站、服務、套件、工具、日誌都沒有監控告警」 | Backup observability coverage:backup job inventory、last success、freshness、offsite、restore drill、Telegram/AwoooP receipt | 已有 backup health exporter / alert rules / Gitea bundle restore dry-run;2026-07-03 runtime 讀回 110 有 88 個 receipt stage requirement、188 有 12 個,`BackupAlertReceiptMetricMissing*` inactive,`BackupAlertReceiptStageMissing` 聚合 pending 110 / 188 各一條 | 補 `/backup/alert-receipts/*.last_success`;再補 Gitea full dump / DB / settings / issues / packages / LFS 與所有工具/log 全量備份監控 | -| 8 | CIR-P0-RBT-008 | P0 | 「每次重啟排查都不一樣,也不知道多久恢復,不符合 SLA」 | 固定化 reboot runbook:fixed triage order、ETA、active blocker、remaining seconds、owner lane、next command | Production scorecard readback 已固定 `status=blocked_reboot_auto_recovery_slo_not_ready`、readiness `60%`、active blockers `8`、primary `fresh_all_host_reboot_event_missing`;12:00 next executable step 固定為 `locate_or_relink_missing_vmx_source_check_mode`,且 locator 結果要求先補 identity / backup evidence | 優先收斂 111 VMX source verified artifact、111 reachability / fresh all-host reboot event、188 startup failed/degraded;不得用不同排查路徑繞過 scorecard | +| 8 | CIR-P0-RBT-008 | P0 | 「每次重啟排查都不一樣,也不知道多久恢復,不符合 SLA」 | 固定化 reboot runbook:fixed triage order、ETA、active blocker、remaining seconds、owner lane、next command | Production scorecard readback 已固定 `status=blocked_reboot_auto_recovery_slo_not_ready`、readiness `60%`、active blockers `10`、primary `reboot_event_required_host_unreachable`;12:34 next executable step 固定為 `locate_or_relink_missing_vmx_source_check_mode`,且 locator 結果要求先補 identity / backup evidence | 優先收斂 111 VMX source verified artifact、111 reachability / fresh all-host reboot event、188 startup failed/degraded;不得用不同排查路徑繞過 scorecard | | 9 | CIR-P0-RBT-009 | P0 | 「所有產品、網站都要是最新版本;版本和數據是否最新要驗證」 | Product freshness/version matrix:source commit、deploy marker、runtime image、public health、data freshness、latest source availability | AWOOOI Gitea `main=bc8d3af3f chore(cd): deploy dacdd90 [skip ci]`;CD `#4558` log 顯示 API/Web build/push wrapper 全部 `rc=0`、deploy marker `dacdd90`、production deploy readback matched。Production SLO route 已讀回最新 API 行為:`active_blocker_count=10`、`runtime_metric_runtime_readback_added_blockers=["windows99_vmware_autostart_config_not_ready","windows99_update_no_auto_reboot_policy_not_ready"]`、`windows99_update_no_auto_reboot_ready=false`。StockPlatform public freshness / ingestion 仍為 `ok`,latest trading date `2026-07-02`。 | 版本層已補到 AWOOOI production;下一步仍要建立全產品 readback 表,且收斂 Windows99 VMX / service / policy / power 與 host reboot-event blockers | | 10 | CIR-P0-GIT-001 | P0 | 「Gitea 儲存庫都不見了?Gitea 沒完整備份嗎?」 | Gitea repository identity + backup proof + restore drill:不能只看 UI visible,要比對 SSH heads、repo path、bundle backup、restore sample | 2026-07-02 production `/api/v1/agents/gitea-repo-bundle-backup-readback` 已 ready:9 expected repos present/ok、missing=0、failed=0、checksum_missing=0、bundle_fresh=true、all_expected_ok=true、sample_restore_dry_run_ok=true;repo bundle / restore dry-run 層已關閉,不是 repo missing。 | 維持每日 bundle backup + restore dry-run monitoring;另補 Gitea full dump / DB / settings / issues / packages / LFS 備份 readback。禁止刪 repo / 改 visibility / 讀 token / restore 到 production | | 11 | CIR-P0-CPU-001 | P0 | 「110 / 188 CPU 負載持續過高,為什麼沒監控告警、沒主動修復」 | Sustained CPU pressure automation:Alertmanager → controller → evidence → service playbook → verifier → KM writeback | 110 已有 `Host110SustainedModeratePressure`、Gitea playbook、Stock/Postgres evidence;188 仍需同級 controller/alerts readback | 下一步接 `postgres_hot_query_or_backup_export_playbook`;並補 188 equivalent readback,不以單次下降結案 | @@ -134,10 +134,10 @@ | KM / PlayBook / RAG / MCP 整合 | 已被列為 P1,不再遺漏 | 建立 work item schema 與 trust writeback 欄位 | | OpenClaw / Gather-style 持續動畫工作室 | route 已存在,已列為 P1 工作項 | 補 production desktop/mobile smoke、AwoooP 導流與截圖證據 | | AI 專業 UI / 非文字牆 cockpit | 已列為 P2 UX 驗收 | 將長文字區塊收斂成 first-viewport cockpit、cards、flow rows 與 expandable details | -| 10 分鐘 reboot auto-recovery SLA | 2026-07-03 12:00 production scorecard 固定 `8` blockers / readiness `60%`,但 `can_claim_all_services_recovered_within_target=false`;primary `fresh_all_host_reboot_event_missing`,Windows99 verify stdout 已收,111 VMX source / guest power 仍 blocked | 收斂 111 VMX source verified artifact、111 reachability、188 startup failed/degraded,再 rerun host probe + reboot-event detector | +| 10 分鐘 reboot auto-recovery SLA | 2026-07-03 12:34 production scorecard 固定 `10` blockers / readiness `60%`,但 `can_claim_all_services_recovered_within_target=false`;primary `reboot_event_required_host_unreachable`,Windows99 locator endpoint ready,111 VMX source / guest power / config / policy 仍 blocked | 收斂 111 VMX source verified artifact、111 reachability、188 startup failed/degraded,再 rerun host probe + reboot-event detector | | AWOOOI latest source production deploy readback | Gitea `main=bc8d3af3f`,deploy marker `dacdd90`;public CD `#4558` log job succeeded / production deploy readback matched;production scorecard 已讀回 latest API 行為並把 Windows99 config / policy blockers 上卷 | 下一步轉回 Windows99 VMX / service / policy / power 與 host reboot-event blockers;不得因 deploy/readback 成功宣稱 10 分鐘 SLO 完成 | | 99 Windows / VMware autostart | Source verifier / parser / API readback / collection packet 已完成;live 99 Verify output 已透過 no-secret collector 收集,VMware services / scheduled task / Windows Update policy ok;VMX locate check 顯示 expected 111 path missing、無 alias / identity target hint、只有一個 generic Ubuntu candidate 需再確認 | 取得 external verified backup artifact 或更強 identity evidence,產生 scoped relink / restore dry-run,重跑 no-secret collector,確認 VM 111/188/120/121/112 running | -| Reboot SLO blocker 收斂 | production scorecard 已固定 `8` blockers / readiness `60%`,source / runtime 已補 per-blocker metric/alert;剩餘主 blocker 是 fresh all-host reboot event + Windows99 111 VMX source / guest power + 111 reachability / 188 degraded | 依具名 blocker 收斂 99/111/188;不得用 route green 或 RDP 可見宣稱 SLA 完成 | +| Reboot SLO blocker 收斂 | production scorecard 已固定 `10` blockers / readiness `60%`,source / runtime 已補 per-blocker metric/alert;剩餘主 blocker 是 reboot event / unreachable + Windows99 111 VMX source / guest power / config / policy + 111 reachability / 188 degraded | 依具名 blocker 收斂 99/111/188;不得用 route green 或 RDP 可見宣稱 SLA 完成 | | 全備份監控告警 coverage | exporter/rule 已有 host/DB/site/service/package/tool/log coverage 與 backup alert receipt requirement;runtime rules 已部署,缺段 alert 聚合成 110 / 188 host-level pending;production receipt markers / full Gitea dump / DB/settings/issues/packages/LFS 尚未全綠 | 補 `/backup/alert-receipts/*.last_success`、Gitea full dump 與 restore drill verifier | | Stock/Postgres hot pressure | 110 live 已導向 Stock/Postgres playbook;尚未完成 hot query / backup export playbook closure | 下一步執行 read-only Stock/Postgres evidence 與 source freshness / query attribution | @@ -154,7 +154,7 @@ ## 下一輪固定執行順序 1. 先確認 Gitea `main` / CD / production readback 是否仍與最新 truth 一致。 -2. 接續目前 active reboot P0:用 12:00 production scorecard 的固定順序收斂 111 VMX source verified artifact、111 reachability、188 startup failed/degraded,再 rerun host probe / reboot-event detector / scorecard。 +2. 接續目前 active reboot P0:用 12:34 production scorecard 的固定順序收斂 111 VMX source verified artifact、111 reachability、188 startup failed/degraded,再 rerun host probe / reboot-event detector / scorecard。 3. 補全 reboot auto-recovery P0:99/110/111/112/120/121/188 reboot detection、10 分鐘 SLO scorecard、Telegram down/up/recovered/SLO missed / active-blocker receipts。 4. 補 99 Windows / VMware autostart P0:99 host 自動啟動 VMware,VM 111/188/120/121/112 自動啟動;同時補 Windows Update no-auto-restart verifier。 5. 補 111 reachability / host boot readback,讓 `host_unreachable_after_reboot` 與 `reboot_event_required_host_unreachable` 有單一固定處理路徑。