From c73f83b99ec41a2e85fd57a2036dbdc64f6ab620 Mon Sep 17 00:00:00 2001 From: Your Name Date: Tue, 30 Jun 2026 23:29:05 +0800 Subject: [PATCH] docs(recovery): record post-push cold start blockers --- docs/LOGBOOK.md | 16 ++++++++++++++++ ...reboot-cold-start-backup-recovery-workplan.md | 8 ++++---- 2 files changed, 20 insertions(+), 4 deletions(-) diff --git a/docs/LOGBOOK.md b/docs/LOGBOOK.md index e48664c4..e2e0a201 100644 --- a/docs/LOGBOOK.md +++ b/docs/LOGBOOK.md @@ -1,3 +1,19 @@ +## 2026-06-30 — 23:27 Post-push cold-start / Harbor / Stock readback + +**照主線修正的問題**: +- `3de828f97 fix(agent): surface harbor cd retry receipt blocker` 已推到 Gitea `main`,隨後同主線 fast-forward 到 `68084470 feat(agent): expose deploy marker receipt input`;本地 worktree 已同步 `gitea-ssh/main`。 +- 推後 public queue 回 `status=blocked_harbor_110_repair_no_matching_runner`,latest CD `#4117 Running`、commit `6808447069422989c547c5edd5e88bcb745dd16b`,Harbor repair `#4115 Waiting`、no-matching label `awoooi-host`,repair jobs 仍 stale/mismatched。 +- Live route probe 仍回 `registry.wooo.work/v2=502`、`192.168.0.110:5000/v2=502`、`harbor.wooo.work/api/v2.0/health=502`、`signoz.wooo.work=502`;不可宣稱全站已恢復。 +- `scripts/reboot-recovery/full-stack-cold-start-check.sh` artifact `/tmp/awoooi-cold-start-after-3de828f97.log` 回 `PASS=67 WARN=5 BLOCKED=4`,結果 `BLOCKED`;blockers 是 110 registry external `/v2`、110 SSH read-only check、K3s registry pull refused by `110:5000`、SignOz TLS/public route。 +- StockPlatform public route / health 為 200,但 `/api/v1/system/freshness` 與 `/api/v1/system/ingestion` 仍回 `status=not_configured`、`blockers=["postgres_not_ready"]`、`latest_trading_date=null`;不可宣稱產品資料最新。 + +**驗證**: +- `pytest apps/api/tests/test_harbor_registry_controlled_recovery_receipt.py apps/api/tests/test_awoooi_priority_work_order_readback_api.py apps/api/tests/test_ai_agent_log_controlled_writeback_executor_readback_api.py ops/runner/test_read_public_gitea_actions_queue.py -q` 通過(41 passed)。 +- `py_compile`、`ruff check`、`guard-gitea-runner-pressure.py --root .`、`git diff --check` 通過。 +- `git ls-remote gitea-ssh refs/heads/main` 回 `6808447069422989c547c5edd5e88bcb745dd16b`。 + +**邊界**:只做 Gitea fast-forward、source/test/doc commit、public queue / HTTP / cold-start readback;未讀 secret / token / `.env` / raw sessions / SQLite / auth;未使用 GitHub / `gh` / GitHub API;未 workflow_dispatch;未 SSH 寫入、未重啟主機、未 restart Docker / Nginx / K3s / DB / firewall。 + ## 2026-06-30 — 23:18 Harbor recovery receipt absorbs live CD 502 retry **照主線修正的問題**: diff --git a/docs/workplans/2026-06-04-reboot-cold-start-backup-recovery-workplan.md b/docs/workplans/2026-06-04-reboot-cold-start-backup-recovery-workplan.md index da42e3db..4e2a28cd 100644 --- a/docs/workplans/2026-06-04-reboot-cold-start-backup-recovery-workplan.md +++ b/docs/workplans/2026-06-04-reboot-cold-start-backup-recovery-workplan.md @@ -15,11 +15,11 @@ | 優先 | 狀態 | 工作項 | 2026-06-30 證據 | 下一步 / 完成條件 | |------|------|--------|------------------|-------------------| -| P0-1 | BLOCKED | 全主機 cold-start / 10 分鐘自動恢復 SLO | 22:55 live cold-start artifact `/tmp/awoooi-cold-start-live-after-ff.log` 回 `PASS=68 WARN=4 BLOCKED=4`;blockers 是 110 registry external `/v2`、110 SSH read-only check、K3s registry pull refused by `110:5000`、SignOz TLS / 502。22:31 SLO scorecard `/tmp/awoooi-reboot-slo-live-20260630-2231-scorecard.json` 仍回 `can_claim_all_services_recovered_within_target=false`;22:28 post-reboot summary `/tmp/awoooi-post-reboot-readiness-20260630-222856/summary.txt` 回 `SERVICE_GREEN=0`、`PRODUCT_DATA_GREEN=0`、`BACKUP_CORE_GREEN=0`、`HOST_188_SERVICE_GREEN=0`。 | 先修第一個 runtime blocker:110 control path / Harbor registry `/v2`。重跑同一 summary / cold-start / SLO scorecard 到 `SERVICE_GREEN=1`、`POST_START_BLOCKED=0`、`PASS` 無 BLOCKED、all-host required observed/reachable 且 `awoooi_reboot_auto_recovery_slo_ready=1`;不可只用 route 200 或 CD `Running` 宣稱恢復。 | +| P0-1 | BLOCKED | 全主機 cold-start / 10 分鐘自動恢復 SLO | 23:27 live cold-start artifact `/tmp/awoooi-cold-start-after-3de828f97.log` 回 `PASS=67 WARN=5 BLOCKED=4`、`Result: BLOCKED`;blockers 是 110 registry external `/v2`、110 SSH read-only check、K3s registry pull refused by `110:5000`、SignOz TLS / public route。22:31 SLO scorecard `/tmp/awoooi-reboot-slo-live-20260630-2231-scorecard.json` 仍回 `can_claim_all_services_recovered_within_target=false`;22:28 post-reboot summary `/tmp/awoooi-post-reboot-readiness-20260630-222856/summary.txt` 回 `SERVICE_GREEN=0`、`PRODUCT_DATA_GREEN=0`、`BACKUP_CORE_GREEN=0`、`HOST_188_SERVICE_GREEN=0`。 | 先修第一個 runtime blocker:110 control path / Harbor registry `/v2`。重跑同一 summary / cold-start / SLO scorecard 到 `SERVICE_GREEN=1`、`POST_START_BLOCKED=0`、`PASS` 無 BLOCKED、all-host required observed/reachable 且 `awoooi_reboot_auto_recovery_slo_ready=1`;不可只用 route 200 或 CD `Running` 宣稱恢復。 | | P0-2 | DONE_THIS_INCIDENT | 使用者可見 502:Tsenyang | `www.tsenyang.com` / `tsenyang.com` 由 502 恢復為 200;188 `tsenyang-website` container running;local `127.0.0.1:3000` 回 200。 | 下次同類 502 先查 release symlink / image / container;不先動 Nginx、DNS、DB、主機重啟。 | -| P0-3 | BLOCKED | StockPlatform data freshness | 22:50 public `/api/v1/system/freshness` 與 `/api/v1/system/ingestion` 回 `status=not_configured`、`blockers=["postgres_not_ready"]`;public route `https://stock.wooo.work/` 為 200 只代表網站可達,不代表資料最新。 | 恢復 110 control path 後,read-only 查 `/home/wooo/stockplatform-v2` compose / DB schema / migration status;禁止 fake freshness、manual DB rows、restore/prune。 | -| P0-4 | BLOCKED | AWOOOI production 版本最新性 | Gitea SSH `main` 最新已到 `a7b79b7b feat(agent): expose harbor receipt input contract`。Public Gitea queue 23:18 讀到 latest CD `#4113 Running`,但 `current_main_cd_inflight_classifier=harbor_registry_public_route_unavailable_pending_retry`、latest registry `/v2` status `502`、Harbor login attempt `8`。production 尚不能證明已部署最新 source。 | 補 deploy marker / runtime SHA / endpoint readback 一致;Harbor `/v2` 恢復前 CD 無法把最新 source 發到 production,未一致前不可宣稱 AWOOOI 最新。 | -| P0-5 | BLOCKED | 110 control path / Harbor registry `/v2` | 23:18 queue readback 回 `status=blocked_harbor_110_repair_no_matching_runner`;Harbor repair workflow `#4112` 仍 `Waiting`,no-matching label `awoooi-host`,jobs API 仍 stale/mismatched `ai-code-review`。Harbor receipt validator 對 live queue 回 `status=blocked_waiting_harbor_controlled_recovery_receipt`,active blockers 含 `gitea_queue_current_cd_harbor_retrying_unavailable`、`gitea_queue_harbor_110_repair_no_matching_runner`、`gitea_queue_harbor_110_repair_jobs_stale_or_mismatched`、`public_registry_v2_verifier_not_green`、`internal_registry_v2_verifier_not_green`。22:55 cold-start 同步證明 110 registry `/v2` blocked、110 SSH read-only check blocked、K3s pull refused by `110:5000`。 | 讓 110-local repair workflow 或 110 console/local script 真正執行 `recover-110-control-path-and-harbor-local.sh --check` / `--apply-all`,並讀回 public/internal `/v2` 為 `200/401`。恢復 SSH read-only command path 後才能驗證 Stock DB、Gitea dump、110 backup completeness。 | +| P0-3 | BLOCKED | StockPlatform data freshness | 23:27 public route / health 為 200,但 `/api/v1/system/freshness` 與 `/api/v1/system/ingestion` 回 `status=not_configured`、`blockers=["postgres_not_ready"]`、`latest_trading_date=null`;網站可達不代表資料最新。 | 恢復 110 control path 後,read-only 查 `/home/wooo/stockplatform-v2` compose / DB schema / migration status;禁止 fake freshness、manual DB rows、restore/prune。 | +| P0-4 | BLOCKED | AWOOOI production 版本最新性 | Gitea SSH `main` 最新已到 `68084470 feat(agent): expose deploy marker receipt input`,且包含 `3de828f97 fix(agent): surface harbor cd retry receipt blocker`。Public Gitea queue 23:27 讀到 latest CD `#4117 Running` for `68084470`;production 尚不能證明已部署最新 source。 | 補 deploy marker / runtime SHA / endpoint readback 一致;Harbor `/v2` 恢復前 CD 無法把最新 source 發到 production,未一致前不可宣稱 AWOOOI 最新。 | +| P0-5 | BLOCKED | 110 control path / Harbor registry `/v2` | 23:27 queue readback 回 `status=blocked_harbor_110_repair_no_matching_runner`;Harbor repair workflow `#4115` 仍 `Waiting`,no-matching label `awoooi-host`,jobs API 仍 stale/mismatched。Live route probe 同步回 public registry `/v2=502`、internal `192.168.0.110:5000/v2=502`、Harbor health `502`、SignOz `502`。Harbor receipt validator 對 live queue 回 `status=blocked_waiting_harbor_controlled_recovery_receipt`,active blockers 含 `gitea_queue_harbor_110_repair_no_matching_runner`、`gitea_queue_harbor_110_repair_jobs_stale_or_mismatched`、`public_registry_v2_verifier_not_green`、`internal_registry_v2_verifier_not_green`。23:27 cold-start 同步證明 110 registry `/v2` blocked、110 SSH read-only check blocked、K3s pull refused by `110:5000`。 | 讓 110-local repair workflow 或 110 console/local script 真正執行 `recover-110-control-path-and-harbor-local.sh --check` / `--apply-all`,並讀回 public/internal `/v2` 為 `200/401`。恢復 SSH read-only command path 後才能驗證 Stock DB、Gitea dump、110 backup completeness。 | | P0-6 | BLOCKED_BACKUP_COMPLETENESS | Gitea repo visibility 與完整備份 | Gitea version API 200;public repo search 只列 4 個 public repo;`stockplatform-v2` public page/API 404,但 internal `git ls-remote` 成功;188 `/home/ollama/backup/110/gitea` 起初為空。已建立 verified emergency bundle `/home/ollama/backup/110/gitea/git-bundles/20260630-190931`:4 個 public/internal repo bundle verify + checksum 成功,`AwoooGo`、`stockplatform-v2`、`vibework` 因 private auth fail-closed。20:18 summary 因 110 `backup-status` 不可讀回,`BACKUP_CORE_GREEN=0`、`DR_ESCROW_BLOCKED=1`、`DR_ESCROW_EVIDENCE_UNKNOWN=1`。 | 188 `gitea_repo_mirror_from_110` subtree metric / alert 已補;下一步仍是恢復 110 SSH command path 後跑正式 `gitea dump`、private repo 非互動備份、repo count、backup-status 與 restore drill readback。unknown 不得當作 backup / DR green。 | | P0-7 | SOURCE_READY_RUNTIME_BLOCKED | 99 VMware / VM autostart | repo 已有 `windows99-vmware-autostart.ps1`;22:05 host probe 讀到 99 ping reachable 但 `boot_id=reachable_unknown_boot` / uptime unknown,111 不可達,112/120/121/188 可讀,188 startup unit failed/degraded。先前只讀 readback 顯示 99 RDP 3389 / SSH 22 可達、WinRM 5985 fail,`administrator@192.168.0.99` SSH publickey denied。 | 恢復 99 可控通道或由 console 套用腳本;完成後讀回 111/188/120/121/112 boot evidence,要求 all-host required observed/reachable 且 99 不再是 unknown uptime。 | | P0-8 | SOURCE_READY_RUNTIME_BLOCKED | 502 maintenance fallback / Telegram / backup alert | L0/L1 fallback runbook、Nginx snippet、reboot / backup alert rules 已在 source;runtime 尚需部署與外部 L1 provider readback。 | L0 以測試 vhost 驗證 `X-AWOOOI-Fallback`;L1 需外部雲端/CDN probe;Telegram 以脫敏 alert receipt 驗證。 |