wooo/awoooi
1
1
Fork 0
Code Issues 2 Pull Requests 1 Actions 2 Packages Projects Releases Wiki Activity

fix(agent99): preserve host112 recovery routing

Browse Source
This commit is contained in:
ogt
2026-07-11 20:09:15 +08:00
parent 43a3a06a3a
commit 55f075f176
2 changed files with 5 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
agent99-telegram-inbox.ps1
View File

@@ -284,8 +284,9 @@ function Get-AgentTelegramAlertSeverity {
$lower = $Text.ToLowerInvariant()
if ($lower -match "resolved|recovered" -or (Test-AgentTextContainsCode $Text @(0x6062, 0x89E3))) { return "resolved" }
if ($lower -match "critical|p0|emergency|fatal|down" -or (Test-AgentTextContainsCode $Text @(0x56B4, 0x91CD, 0x7DCA, 0x6025))) { return "critical" }
if ($lower -match "warning|warn|degraded|readback_failed" -or (Test-AgentTextContainsCode $Text @(0x544A, 0x8B66, 0x7570, 0x5E38))) { return "warning" }
if ($lower -match "missing_key|e_ai_missing_key|application_ai_config_error|locale guard|ai support") { return "warning" }
if ($lower -match "failed" -or $Text -match "失敗") { return "critical" }
if ($lower -match "warning|warn|degraded|readback_failed|missing_key|locale guard|ai support" -or (Test-AgentTextContainsCode $Text @(0x544A, 0x8B66, 0x7570, 0x5E38))) { return "warning" }
return "warning"
}

6
ops/monitoring/alerts-unified.yml
View File

@@ -260,7 +260,7 @@ groups:
automation_mode: controlled_apply
annotations:
summary: "112 開機完成讀回中斷"
description: "110 無法取得 112 的 forced-command readiness；Ping 或 SSH port 可達不代表 Kali 圖形登入、VMware Tools 與 Wazuh 已恢復。"
description: "reboot-auto-recovery 無法取得 112 的 forced-command readiness；Ping 或 SSH port 可達不代表 Kali 圖形登入、VMware Tools 與安全服務已恢復。"
runbook: "Agent99 先驗證 112 VMX process 與 SSH，再恢復限定的 110→112 forced-command 公鑰通道；不得因 UI 顯示 Powered off 就重複 power on。"
- alert: Host112GuestNotReady
@@ -278,8 +278,8 @@ groups:
automation_mode: controlled_apply
annotations:
summary: "112 尚未真正完成開機恢復"
description: "112 的 Console、VMware Tools、Wazuh 服務或 recovery timer 至少一項未通過；Agent99 應執行 allowlisted guest recovery 並以 forced-command readback 驗證。"
runbook: "執行 sudo -n /usr/local/sbin/awoooi-host112-guest-readiness --apply；僅允許補起 graphical.target、LightDM、open-vm-tools 與 cooldown-guarded Wazuh Indexer，不得 reboot、reset VM 或讀取 secret。"
description: "reboot-auto-recovery 的 112 Console、VMware Tools、安全服務或 recovery timer 至少一項未通過；Agent99 應執行 allowlisted guest recovery 並以 forced-command readback 驗證。"
runbook: "執行 sudo -n /usr/local/sbin/awoooi-host112-guest-readiness --apply；僅允許補起 graphical.target、LightDM、open-vm-tools 與 cooldown-guarded Indexer，不得 reboot、reset VM 或讀取 secret。"
- alert: RebootAutoRecoverySLOExporterStale
expr: time() - awoooi_reboot_auto_recovery_slo_last_run_timestamp{scope="99_110_111_112_120_121_188"} > 600