非Spring Boot的Spring项目如何通过Micrometer采集REST端点指标

核心实现思路 #

由于非Spring Boot环境没有默认的HTTP指标自动配置能力，我们可以通过Spring MVC原生的拦截器机制实现全局埋点，无需修改现有REST接口代码即可完成全量请求的指标采集。

具体实现步骤 #

• 步骤1：补充依赖（若已传递引入可跳过）
  如果你项目中没有自动引入micrometer核心包，可补充如下依赖：

  <dependency>
      <groupId>io.micrometer</groupId>
      <artifactId>micrometer-core</artifactId>
      <version>1.7.5</version>
  </dependency>
  

• 步骤2：实现HTTP指标采集拦截器
  自定义拦截器实现请求耗时、请求次数、状态码等维度的统计：

  import io.micrometer.core.instrument.MeterRegistry;
  import io.micrometer.core.instrument.Timer;
  import org.springframework.web.servlet.HandlerInterceptor;
  import javax.servlet.http.HttpServletRequest;
  import javax.servlet.http.HttpServletResponse;
  import java.util.concurrent.TimeUnit;
  
  @Component
  public class HttpMetricsInterceptor implements HandlerInterceptor {
      private static final String REQUEST_START_TIME_ATTR = "http_request_start_time";
      private final MeterRegistry meterRegistry;
  
      public HttpMetricsInterceptor(MeterRegistry meterRegistry) {
          this.meterRegistry = meterRegistry;
      }
  
      @Override
      public boolean preHandle(HttpServletRequest request, HttpServletResponse response, Object handler) throws Exception {
          request.setAttribute(REQUEST_START_TIME_ATTR, System.currentTimeMillis());
          return true;
      }
  
      @Override
      public void afterCompletion(HttpServletRequest request, HttpServletResponse response, Object handler, Exception ex) throws Exception {
          Long startTime = (Long) request.getAttribute(REQUEST_START_TIME_ATTR);
          if (startTime == null) {
              return;
          }
          long duration = System.currentTimeMillis() - startTime;
          // 优先取Spring匹配的接口模板路径，避免动态参数导致指标维度爆炸
          String uri = (String) request.getAttribute(org.springframework.web.servlet.HandlerMapping.BEST_MATCHING_PATTERN_ATTRIBUTE);
          if (uri == null) {
              uri = request.getRequestURI();
          }
          // 记录请求指标，自带计数、耗时统计能力
          Timer.builder("http.server.requests")
                  .tag("method", request.getMethod())
                  .tag("status", String.valueOf(response.getStatus()))
                  .tag("uri", uri)
                  .description("HTTP服务端请求指标")
                  .register(meterRegistry)
                  .record(duration, TimeUnit.MILLISECONDS);
      }
  }
  

  注意：不要直接将带动态参数的实际请求路径作为uri标签值，否则会导致指标基数过大，大幅提升Prometheus存储和查询压力
• 步骤3：注册拦截器到Spring MVC上下文
  将自定义拦截器注册到全局拦截链，使其生效：

  import org.springframework.context.annotation.Configuration;
  import org.springframework.web.servlet.config.annotation.InterceptorRegistry;
  import org.springframework.web.servlet.config.annotation.WebMvcConfigurer;
  
  @Configuration
  public class WebMetricsConfig implements WebMvcConfigurer {
      private final HttpMetricsInterceptor httpMetricsInterceptor;
  
      public WebMetricsConfig(HttpMetricsInterceptor httpMetricsInterceptor) {
          this.httpMetricsInterceptor = httpMetricsInterceptor;
      }
  
      @Override
      public void addInterceptors(InterceptorRegistry registry) {
          // 可根据需求调整拦截路径范围，此处拦截所有请求
          registry.addInterceptor(httpMetricsInterceptor).addPathPatterns("/**");
      }
  }
  

  如果你使用的是Spring 4及更早版本，可继承WebMvcConfigurerAdapter完成相同配置。
• 步骤4：验证指标
  项目重启后，调用现有REST接口后访问Prometheus指标拉取端点，即可看到http_server_requests_seconds_count（请求次数）、http_server_requests_seconds_sum（总耗时）等相关指标，自带method、status、uri多个维度标签。

磁盘空间指标补充 #

如果需要采集磁盘空间指标，可在项目启动时将磁盘指标绑定到MeterRegistry即可自动采集：

import io.micrometer.core.instrument.MeterRegistry;
import io.micrometer.core.instrument.binder.system.DiskSpaceMetrics;
import javax.annotation.PostConstruct;
import java.io.File;

@Component
public class ExtraMetricsBinder {
    private final MeterRegistry meterRegistry;

    public ExtraMetricsBinder(MeterRegistry meterRegistry) {
        this.meterRegistry = meterRegistry;
    }

    @PostConstruct
    public void bindMetrics() {
        // 可替换为你需要监控的磁盘挂载路径
        new DiskSpaceMetrics(new File("/")).bindTo(meterRegistry);
    }
}

内容的提问来源于stack exchange，提问作者Sachin Chandra