From 0a7edbf3ce20f39632a2d85855644094b568d4a2 Mon Sep 17 00:00:00 2001 From: Your Name Date: Thu, 2 Jul 2026 20:17:53 +0800 Subject: [PATCH] docs(alerts): record alertmanager delivery proof [metadata-only] --- docs/LOGBOOK.md | 3 +++ 1 file changed, 3 insertions(+) diff --git a/docs/LOGBOOK.md b/docs/LOGBOOK.md index d8f8ba156..f7cce8237 100644 --- a/docs/LOGBOOK.md +++ b/docs/LOGBOOK.md @@ -13,6 +13,9 @@ - `bash -n scripts/ops/deploy-alertmanager-config.sh`：通過。 - `python3 ops/runner/guard-gitea-runner-pressure.py --root .`：`GITEA_RUNNER_PRESSURE_GUARD_OK ... auto_branch_events_on_110=0 generic_runner_labels=0`。 - live readback：Alertmanager ready HTTP `200`；public webhook probe HTTP `200` / `No firing alerts to process`；Prometheus 仍讀回 `Host110SustainedModeratePressure` firing `3`、`HostRebootEventDetected` firing `1`、`RebootAutoRecoverySLOMissed` firing `1`。 +- Alertmanager 容器內 config 讀回新 URL count `1`、舊 `192.168.0.125:32334` count `0`；容器內 direct firing payload probe 回 `success=true` / `alert-20260702201413` / `TYPE-1 純資訊告警已通知 (no LLM)`。 +- Alertmanager notifier controlled probe 後，`alertmanager_notifications_total{integration="webhook"}` 從 `177` 增為 `179`，`alertmanager_notifications_failed_total{integration="webhook",reason="other"}` 維持 `1` 未增加；`awoooi_alert_chain_last_success_timestamp{source="alertmanager"}` age 讀回 `0` 秒。 +- 仍需另列 metrics gap：`awoooi_webhook_requests_total{source="alertmanager"}` 在本輪讀回無 series，不能拿它當 delivery 成功指標；後續要補 webhook request counter exposure / scrape contract。 **仍維持**： - 未讀 Telegram token / secret / `.env` / raw sessions / SQLite / auth；未呼叫 direct Bot API；未 workflow_dispatch；未重啟主機 / VM / Docker daemon / Nginx / K3s / DB / firewall；未 fake DB rows；未宣稱 10 分鐘全恢復完成。