xx

alertmanager/aserver/alertmanager-webhook-test.yaml

@@ -0,0 +1,27 @@
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: alertmanager-config
+  namespace: default
+data:
+  alertmanager.yml: |
+    route:
+      group_by: ['alertname', 'service']
+      group_wait: 5s
+      group_interval: 5s
+      repeat_interval: 15m
+      receiver: 'aserver-webhook-receiver'
+      routes:
+      - match:
+          service: aserver
+          alert_type: offlines
+        receiver: 'aserver-webhook-receiver'
+        group_wait: 5s
+        group_interval: 5s
+        repeat_interval: 15m
+    
+    receivers:
+    - name: 'aserver-webhook-receiver'
+      webhook_configs:
+      - url: 'http://alert:8080/webhook/alert'
+        send_resolved: true

prometheus/aserver/aserver-offline-alert.yaml

@@ -0,0 +1,22 @@
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: aserver-offline-alert
+  namespace: default
+data:
+  aserver-offline-alert.yml: |
+    groups:
+    - name: aserver-offline-alerts
+      rules:
+      - alert: AServerOffline
+        expr: up{kubernetes_pod_name=~"aserver.*"} == 0
+        for: 1m
+        labels:
+          severity: critical
+          service: aserver
+          alert_type: offline
+        annotations:
+          summary: "AServer 服务离线"
+          description: "AServer Pod {{ $labels.kubernetes_pod_name }} 已离线超过1分钟"
+          runbook_url: "https://docs.example.com/aserver-troubleshooting"
+          action: "请检查 Pod 状态和日志"

prometheus/aserver/install-aserver-alerts.sh

@@ -0,0 +1,41 @@
+#!/bin/bash
+set -euo pipefail
+
+echo "开始安装 AServer 告警规则到 Prometheus..."
+
+echo "1. 应用 AServer 离线告警规则..."
+kubectl apply -f aserver-offline-alert.yaml
+
+echo "2. 重启 Prometheus 以加载新的告警规则..."
+kubectl rollout restart deployment/prometheus
+
+echo "等待 Prometheus 重启..."
+kubectl wait --for=condition=available --timeout=300s deployment/prometheus
+
+echo ""
+echo "=========================================="
+echo "AServer 告警规则安装完成！"
+echo "=========================================="
+echo ""
+echo "告警规则："
+echo "- AServerOffline: aserver离线告警 (1分钟后触发)"
+echo "  - 严重程度: critical"
+echo "  - 触发条件: up{kubernetes_pod_name=~\"aserver.*\"} == 0"
+echo "  - 持续时间: 1分钟"
+echo ""
+echo "访问地址："
+echo "- Prometheus Alerts: https://prom-ops.t-aaron.com/alerts"
+echo ""
+echo "验证方法："
+echo "1. 访问 Prometheus Alerts 页面"
+echo "2. 查找 'AServerOffline' 告警规则"
+echo "3. 状态应该显示为 'Inactive' (正常状态)"
+echo ""
+echo "测试告警："
+echo "1. 删除 aserver Pod: kubectl delete pod -l app=aserver"
+echo "2. 等待1分钟后查看告警状态变为 'Pending' 或 'Firing'"
+echo "3. 重新部署 aserver 后告警应自动恢复"
+echo ""
+echo "注意：此脚本只安装 Prometheus 告警规则"
+echo "如需钉钉通知，请运行 Alertmanager 的安装脚本"
+echo "=========================================="

prometheus/prometheus-rules-configmap.yaml

@@ -54,3 +54,17 @@ data:
         annotations:
           summary: "Prometheus configuration reload failed"
           description: "Prometheus configuration reload has been failing for more than 5 minutes."
+
+    - name: aserver-alerts
+      rules:
+      - alert: AServerOffline
+        expr: absent(up{kubernetes_pod_name=~"aserver.*"})
+        for: 1m
+        labels:
+          severity: critical
+          service: aserver
+          alert_type: offline
+        annotations:
+          summary: "AServer 服务离线"
+          description: "AServer Pod 不存在或已离线超过1分钟"
+          action: "请检查 Pod 状态和日志"