AWS SDK Java v2自定义指标收集器等效实现咨询（Spring Boot+Micrometer）

Java SDK v2不再提供AwsSdkMetrics这类全局静态配置方式，而是通过拦截器（Interceptor）和Micrometer集成模块实现自动指标收集，完全适配Spring Boot环境，无需手动封装每个客户端调用。

1. 添加依赖 #

首先引入AWS SDK v2的Micrometer集成依赖（以Maven为例）：

<dependency>
    <groupId>software.amazon.awssdk</groupId>
    <artifactId>micrometer-core</artifactId>
    <version>${aws.sdk.version}</version>
</dependency>

Gradle依赖配置：

implementation "software.amazon.awssdk:micrometer-core:${awsSdkVersion}"

2. 全局配置AWS客户端指标拦截器 #

方式一：逐个客户端绑定指标发布器 #

通过Spring配置类为每个AWS客户端绑定Micrometer指标发布器：

import io.micrometer.core.instrument.MeterRegistry;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
import software.amazon.awssdk.metrics.MetricPublisher;
import software.amazon.awssdk.metrics.micrometer.MicrometerMetricPublisher;
import software.amazon.awssdk.regions.Region;
import software.amazon.awssdk.services.s3.S3Client;
import software.amazon.awssdk.services.dynamodb.DynamoDbClient;

@Configuration
public class AwsClientConfig {

    // 创建全局Micrometer指标发布器
    @Bean
    public MetricPublisher micrometerMetricPublisher(MeterRegistry meterRegistry) {
        return MicrometerMetricPublisher.create(meterRegistry);
    }

    // 配置S3客户端并绑定指标拦截器
    @Bean
    public S3Client s3Client(MetricPublisher metricPublisher) {
        return S3Client.builder()
                .region(Region.US_EAST_1)
                .overrideConfiguration(config -> config.addMetricPublisher(metricPublisher))
                .build();
    }

    // 配置DynamoDB客户端并绑定指标拦截器
    @Bean
    public DynamoDbClient dynamoDbClient(MetricPublisher metricPublisher) {
        return DynamoDbClient.builder()
                .region(Region.US_EAST_1)
                .overrideConfiguration(config -> config.addMetricPublisher(metricPublisher))
                .build();
    }
}

方式二：全局默认配置（所有客户端自动生效） #

如果希望所有AWS客户端自动应用指标收集，无需逐个配置，可以设置全局默认客户端配置：

import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
import org.springframework.context.annotation.PostConstruct;
import software.amazon.awssdk.core.client.config.ClientOverrideConfiguration;
import software.amazon.awssdk.core.client.config.SdkClientConfiguration;
import software.amazon.awssdk.core.client.config.SdkClientOption;
import software.amazon.awssdk.metrics.MetricPublisher;

@Configuration
public class AwsGlobalConfig {

    @Bean
    public ClientOverrideConfiguration globalClientOverrideConfiguration(MetricPublisher metricPublisher) {
        return ClientOverrideConfiguration.builder()
                .addMetricPublisher(metricPublisher)
                .build();
    }

    @PostConstruct
    public void setGlobalAwsConfig(ClientOverrideConfiguration overrideConfig) {
        SdkClientConfiguration globalConfig = SdkClientConfiguration.builder()
                .option(SdkClientOption.CLIENT_OVERRIDE_CONFIG, overrideConfig)
                .build();
        // 全局设置所有AWS客户端的默认配置
        software.amazon.awssdk.core.client.builder.SdkClientBuilder.setConfiguration(globalConfig);
    }
}

3. 验证指标收集 #

启动Spring Boot应用后，访问Prometheus端点（默认/actuator/prometheus），可以看到自动生成的AWS相关指标，例如：

• aws_sdk_requests_total：按服务、操作、状态码区分的总请求数
• aws_sdk_request_duration_seconds：请求耗时分布统计
• aws_sdk_request_retries_total：请求重试次数统计

这些指标会自动注入Micrometer的MeterRegistry，被Prometheus抓取后即可在Grafana中可视化。

关键说明 #

• Java SDK v2采用依赖注入式配置，替代了v1的全局静态模式，避免了静态初始化的副作用
• Micrometer集成模块已封装核心指标收集逻辑，无需自定义Collector类
• 若需扩展自定义指标，可实现MetricPublisher接口，或在Micrometer层面添加自定义标签

内容的提问来源于stack exchange，提问作者Miles Elam