SpringBoot3中LoggingMeterRegistry配置及MetricFilter自定义问题

1. 通过 application.yaml 配置 step 等参数 #

直接new LoggingMeterRegistry()不会自动读取配置文件属性，需要借助Spring Boot提供的LoggingProperties绑定配置，让Bean感知yaml中的参数：

步骤1：修改Bean定义，注入LoggingProperties #

import io.micrometer.core.instrument.Clock;
import io.micrometer.core.instrument.logging.LoggingMeterRegistry;
import io.micrometer.core.instrument.logging.LoggingRegistryConfig;
import org.springframework.boot.actuate.autoconfigure.metrics.export.logging.LoggingProperties;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;

@Configuration
public class MeterRegistryConfig {
    @Bean
    LoggingMeterRegistry loggingMeterRegistry(LoggingProperties loggingProperties) {
        LoggingRegistryConfig config = loggingProperties.getExport().getLogging();
        return new LoggingMeterRegistry(config, Clock.SYSTEM);
    }
}

步骤2：在 application.yaml 中配置参数 #

使用management.metrics.export.logging前缀配置相关属性：

management:
  metrics:
    export:
      logging:
        enabled: true
        step: 30s  # 指标输出间隔，默认值为1分钟（1m）
        log-level: INFO  # 日志输出级别，可选TRACE/DEBUG/INFO/WARN/ERROR
        descriptions: false  # 是否输出指标描述，默认false

2. 注册自定义 MetricFilter 过滤输出 #

你提到的com.codahale.metrics.MetricFilter是Dropwizard Metrics的API，而Micrometer自身使用io.micrometer.core.instrument.MeterFilter接口，需要先将Dropwizard的Filter适配为Micrometer的Filter，再添加到LoggingMeterRegistry中：

示例代码：适配Dropwizard MetricFilter并注册 #

import com.codahale.metrics.MetricFilter;
import io.micrometer.core.instrument.Clock;
import io.micrometer.core.instrument.MeterFilter;
import io.micrometer.core.instrument.logging.LoggingMeterRegistry;
import io.micrometer.core.instrument.logging.LoggingRegistryConfig;
import org.springframework.boot.actuate.autoconfigure.metrics.export.logging.LoggingProperties;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;

import java.util.Collections;

@Configuration
public class MeterRegistryConfig {
    @Bean
    LoggingMeterRegistry loggingMeterRegistry(LoggingProperties loggingProperties) {
        // 自定义Dropwizard MetricFilter：仅保留名称包含"http"的指标
        MetricFilter dropwizardFilter = (metricName, metric) -> metricName.contains("http");

        // 将Dropwizard Filter适配为Micrometer MeterFilter
        MeterFilter micrometerFilter = MeterFilter.dropFiltered(
            meterId -> !dropwizardFilter.matches(meterId.getName(), null)
        );

        LoggingRegistryConfig config = loggingProperties.getExport().getLogging();
        // 将自定义Filter传入LoggingMeterRegistry构造方法
        return new LoggingMeterRegistry(config, Clock.SYSTEM, Collections.singletonList(micrometerFilter));
    }
}

更简便的方式：直接使用Micrometer原生MeterFilter #

如果不需要依赖Dropwizard的Filter，直接用Micrometer的API更简洁：

@Bean
LoggingMeterRegistry loggingMeterRegistry(LoggingProperties loggingProperties) {
    // 自定义Filter：拒绝所有以"jvm."开头的指标
    MeterFilter customFilter = MeterFilter.denyNameStartsWith("jvm.");

    LoggingRegistryConfig config = loggingProperties.getExport().getLogging();
    return new LoggingMeterRegistry(config, Clock.SYSTEM, Collections.singletonList(customFilter));
}

内容的提问来源于stack exchange，提问作者Antonio Petricca