monitoring-setuplisted

# 监控接入 > 配置 Prometheus + Grafana + 告警。 ## 执行步骤 ### 第 1 步：暴露 Prometheus 端点 ```yaml management: endpoints: web: exposure: include: health,info,metrics,prometheus endpoint: prometheus: enabled: true ``` ### 第 2 步：配置 Prometheus 抓取 ```yaml # prometheus.yml scrape_configs: - job_name: 'user-service' metrics_path: '/actuator/prometheus' static_configs: - targets: ['user-service:8080'] ``` ### 第 3 步：配置告警规则 ```yaml # alert-rules.yml groups: - name: service rules: - alert: ServiceDown expr: up == 0 for: 1m labels: severity: critical ``` ### 第 4 步：Grafana Dashboard 导入或创建 Dashboard。 ### 第 5 步：验证 ```bash # 检查端点 curl http://localhost:8080/actuator/prometheus # 检查 Prometheus 抓取 curl http://prometheus:9090/api/v1/targets ``` ## 关键约束 - ✅ **MUST** 暴露 `/actuator/prometheus` - ✅ **MUST** 配置关键告警 - ❌ **MUST NOT** 在生产环境 100% trace 采样 ## 关联 - Wiki：`wiki/_common/observability.md` - 相关：`deployment-verification`