Prometheus + Grafana + Micrometer 监控方案详解

最新推荐文章于 2025-10-02 10:42:58 发布

原创最新推荐文章于 2025-10-02 10:42:58 发布 · 1.2k 阅读

17 ·

CC 4.0 BY-SA版权

文章标签：

#prometheus #grafana

分布式专栏收录该内容

34 篇文章

订阅专栏

这套组合是当前Java生态中最流行的监控解决方案之一，特别适合云原生环境下的微服务应用监控。下面我将从技术实现到最佳实践进行全面解析。

一、技术栈组成与协作

1. 组件分工

组件	角色	关键能力
Micrometer	应用指标门面(Facade)	统一指标采集API，对接多种监控系统
Prometheus	时序数据库+采集器	指标存储、查询、告警规则处理
Grafana	可视化平台	仪表盘展示、数据可视化分析

2. 数据流动

二、Micrometer 集成实践

1. Spring Boot 配置

Maven依赖：

<dependency>
    <groupId>io.micrometer</groupId>
    <artifactId>micrometer-registry-prometheus</artifactId>
</dependency>
<dependency>
    <groupId>org.springframework.boot</groupId>
    <artifactId>spring-boot-starter-actuator</artifactId>
</dependency>

application.yml配置：

management:
  endpoints:
    web:
      exposure:
        include: health,info,prometheus
  metrics:
    export:
      prometheus:
        enabled: true
    tags:
      application: ${spring.application.name} # 统一添加应用标签

2. 自定义指标示例

业务指标采集：

@Service
public class OrderService {
    private final Counter orderCounter;
    private final Timer orderProcessingTimer;

    public OrderService(MeterRegistry registry) {
        // 创建计数器
        orderCounter = Counter.builder("orders.total")
            .description("Total number of orders")
            .tag("type", "online")
            .register(registry);
            
        // 创建计时器
        orderProcessingTimer = Timer.builder("orders.processing.time")
            .description("Order processing time")
            .publishPercentiles(0.5, 0.95) // 50%和95%分位
            .register(registry);
    }

    public void processOrder(Order order) {
        // 方法1: 手动计时
        long start = System.currentTimeMillis();
        try {
            // 业务逻辑...
            orderCounter.increment();
        } finally {
            long duration = System.currentTimeMillis() - start;
            orderProcessingTimer.record(duration, TimeUnit.MILLISECONDS);
        }
        
        // 方法2: 使用Lambda自动计时
        orderProcessingTimer.record(() -> {
            // 业务逻辑...
            orderCounter.increment();
        });
    }
}

三、Prometheus 配置优化

1. 抓取配置示例

# prometheus.yml
global:
  scrape_interval: 15s
  evaluation_interval: 30s

scrape_configs:
  - job_name: 'spring-apps'
    metrics_path: '/actuator/prometheus'
    scrape_interval: 10s # 对应用更频繁采集
    static_configs:
      - targets: ['app1:8080', 'app2:8080']
    relabel_configs:
      - source_labels: [__address__]
        target_label: instance
      - source_labels: [__meta_service_name]
        target_label: service

2. 关键优化参数

存储配置：

# 控制块存储行为
storage:
  tsdb:
    retention: 15d # 数据保留时间
    out_of_order_time_window: 1h # 允许乱序数据窗口

# 限制内存使用
query:
  lookback-delta: 5m
  max-concurrency: 20

四、Grafana 仪表盘设计

1. 核心监控仪表盘

JVM监控面板配置：

Panel 1: Heap Memory Usage
Query: sum(jvm_memory_used_bytes{area="heap"}) by (instance) / sum(jvm_memory_max_bytes{area="heap"}) by (instance)
Visualization: Time series with % unit

Panel 2: GC Pause Time
Query: rate(jvm_gc_pause_seconds_sum[1m])
Visualization: Heatmap

Panel 3: Thread States
Query: jvm_threads_states_threads{instance=~"$instance"}
Visualization: Stacked bar chart

2. 业务指标可视化

订单业务看板：

{
  "panels": [
    {
      "title": "Orders per Minute",
      "targets": [{
        "expr": "rate(orders_total[1m])",
        "legendFormat": "{{instance}}"
      }],
      "type": "graph",
      "yaxes": [{"format": "ops"}]
    },
    {
      "title": "Processing Time (95%)",
      "targets": [{
        "expr": "histogram_quantile(0.95, rate(orders_processing_time_seconds_bucket[1m]))",
        "legendFormat": "P95"
      }],
      "type": "stat",
      "unit": "s"
    }
  ]
}

五、生产环境最佳实践

1. 指标命名规范

类型	前缀	示例
计数器	`_total`	`http_requests_total`
计量器	`_current`	`queue_size_current`
计时器	`_seconds`	`api_latency_seconds`
分布概要	`_summary`	`response_size_summary`

2. 标签使用原则

避免高基数标签：如用户ID、订单号等
统一标签命名：团队内保持一致（如env vs environment）
重要维度标记：region、az、service_version等

3. 资源优化技巧

Micrometer配置：

@Bean
MeterRegistryCustomizer<MeterRegistry> metricsCommonTags() {
    return registry -> registry.config()
        .meterFilter(
            MeterFilter.deny(id -> {
                // 过滤不需要的指标
                return id.getName().startsWith("jvm_classes");
            }))
        .commonTags("region", System.getenv("AWS_REGION"));
}

Prometheus资源限制：

# 容器部署时设置资源限制
resources:
  limits:
    memory: 8Gi
  requests:
    cpu: 2
    memory: 4Gi

六、高级功能实现

1. 自定义Collector

public class CustomMetricsCollector extends Collector {
    @Override
    public List<MetricFamilySamples> collect() {
        List<MetricFamilySamples> samples = new ArrayList<>();
        
        // 添加自定义指标
        samples.add(new MetricFamilySamples(
            "custom_metric",
            Type.GAUGE,
            "Custom metric description",
            Collections.singletonList(
                new MetricFamilySamples.Sample(
                    "custom_metric",
                    List.of("label1"),
                    List.of("value1"),
                    getCurrentValue()
                )
            )));
        
        return samples;
    }
}

// 注册Collector
new CustomMetricsCollector().register();

2. 告警规则示例

groups:
- name: application-alerts
  rules:
  - alert: HighErrorRate
    expr: rate(http_server_requests_errors_total[5m]) / rate(http_server_requests_total[5m]) > 0.05
    for: 10m
    labels:
      severity: critical
    annotations:
      summary: "High error rate on {{ $labels.instance }}"
      description: "Error rate is {{ $value }}"
      
  - alert: GCTooLong
    expr: rate(jvm_gc_pause_seconds_sum[1h]) > 0.1
    labels:
      severity: warning