Spring Cloud GateWay基于nacos如何去做灰度发布

本文基于 ReactiveLoadBalancerClientFilter使用RoundRobinLoadBalancer

灰度发布

灰度发布，又称为金丝雀发布，是一种新旧版本平滑发布的方式。在上面可以对同一个API进行两个版本 的内容，由一部分用户先行体验，如无问题，逐步扩大发布范围
​​​​​

本文将讲述如何基于基于nacos的matedata与Ribbon如何去做灰度发布

重点知识

Spring Cloud Gateway两种负载均衡器

官网说明两种负载均衡器
Gateway有两种客户端负载均衡器，LoadBalancerClientFilter和ReactiveLoadBalancerClientFilter。
LoadBalancerClientFilter使用一个Ribbon的阻塞式LoadBalancerClient，Gateway建议使用ReactiveLoadBalancerClientFilter。
可以通过设置spring.cloud.loadbalancer.ribbon.enabled=false，切换到ReactiveLoadBalancerClientFilter。无论使用Ribbon还是LoadBalancer，在Route中配置的lb是一样的

本节采用 ReactiveLoadBalancerClientFilter 进行设置

采用ReactiveLoadBalancerClientFilter使用RoundRobinLoadBalancer

灰度发布服务器选择 简单示意图

Client —-> gateway —-> GlobalFilter 拦截 选择一个灰度发布服务器 如果没有灰度服务则选取正常服务器 —->转发到服务

nacos的matedata

我们在向 Nacos Server 进行服务注册的时候往往会附加一些 metadata ，可以参考官方文档中 ​​Dubbo 融合 Nacos 成为注册中心​​​ 章节。
充分利用好服务实例的 metadata ，可以衍生出许多有意思的实践。
完全可以把相关内容放进 metadata 中，好比说版本号，特性名等等

然后再根据负载均衡路由到不同的服务

spring.cloud.nacos.discovery.metadata.version=1.15
spring.cloud.nacos.discovery.metadata.advance=true

​​​​

​​​​

准备工作

nacos 部署
gateway 部署 -可以参考
部署两台服务A

开始

跟踪代码

public Mono<Void> filter(ServerWebExchange exchange, GatewayFilterChain chain) {
    URI url = (URI)exchange.getAttribute(ServerWebExchangeUtils.GATEWAY_REQUEST_URL_ATTR);
    String schemePrefix = (String)exchange.getAttribute(ServerWebExchangeUtils.GATEWAY_SCHEME_PREFIX_ATTR);
    if (url != null && ("lb".equals(url.getScheme()) || "lb".equals(schemePrefix))) {
        ServerWebExchangeUtils.addOriginalRequestUrl(exchange, url);
        if (log.isTraceEnabled()) {
            log.trace(ReactiveLoadBalancerClientFilter.class.getSimpleName() + " url before: " + url);
        }

        return this.choose(exchange).doOnNext((response) -> {
            if (!response.hasServer()) {
                throw NotFoundException.create(this.properties.isUse404(), "Unable to find instance for " + url.getHost());
            } else {
                ServiceInstance retrievedInstance = (ServiceInstance)response.getServer();
                URI uri = exchange.getRequest().getURI();
                String overrideScheme = retrievedInstance.isSecure() ? "https" : "http";
                if (schemePrefix != null) {
                    overrideScheme = url.getScheme();
                }

                DelegatingServiceInstance serviceInstance = new DelegatingServiceInstance(retrievedInstance, overrideScheme);
                URI requestUrl = this.reconstructURI(serviceInstance, uri);
                if (log.isTraceEnabled()) {
                    log.trace("LoadBalancerClientFilter url chosen: " + requestUrl);
                }

                exchange.getAttributes().put(ServerWebExchangeUtils.GATEWAY_REQUEST_URL_ATTR, requestUrl);
            }
        }).then(chain.filter(exchange));
    } else {
        return chain.filter(exchange);
    }
}

protected URI reconstructURI(ServiceInstance serviceInstance, URI original) {
    return LoadBalancerUriTools.reconstructURI(serviceInstance, original);
}

private Mono<Response<ServiceInstance>> choose(ServerWebExchange exchange) {
    URI uri = (URI)exchange.getAttribute(ServerWebExchangeUtils.GATEWAY_REQUEST_URL_ATTR);
    ReactorLoadBalancer<ServiceInstance> loadBalancer = (ReactorLoadBalancer)this.clientFactory.getInstance(uri.getHost(), ReactorServiceInstanceLoadBalancer.class);
    if (loadBalancer == null) {
        throw new NotFoundException("No loadbalancer available for " + uri.getHost());
    } else {
        return loadBalancer.choose(this.createRequest());
    }
}

@SuppressWarnings("rawtypes")
@Override
// see original
// https://github.com/Netflix/ocelli/blob/master/ocelli-core/
// src/main/java/netflix/ocelli/loadbalancer/RoundRobinLoadBalancer.java
public Mono<Response<ServiceInstance>> choose(Request request) {
  // TODO: move supplier to Request?
  // Temporary conditional logic till deprecated members are removed.
  if (serviceInstanceListSupplierProvider != null) {
    ServiceInstanceListSupplier supplier = serviceInstanceListSupplierProvider
        .getIfAvailable(NoopServiceInstanceListSupplier::new);
    return supplier.get().next().map(this::getInstanceResponse);
  }
  ServiceInstanceSupplier supplier = this.serviceInstanceSupplier
      .getIfAvailable(NoopServiceInstanceSupplier::new);
  return supplier.get().collectList().map(this::getInstanceResponse);
}

private Response<ServiceInstance> getInstanceResponse(
    List<ServiceInstance> instances) {
  if (instances.isEmpty()) {
    log.warn("No servers available for service: " + this.serviceId);
    return new EmptyResponse();
  }
  // TODO: enforce order?
  int pos = Math.abs(this.position.incrementAndGet());

  ServiceInstance instance = instances.get(pos % instances.size());

  return new DefaultResponse(instance);
}

通过代码跟踪 ReactiveLoadBalancerClientFilter 与 RoundRobinLoadBalancer 可以发现，最终 我们只需要对 getInstanceResponse 进行改造 即可满足所有需要

动手！

开始修改代码

我们只需要新增一个 GlobalFilter 在 AdvanceReactiveLoadBalancerClientFilter 执行之前 ，并且对LoadBalancer 的getInstanceResponse 做一下稍微改造就OK了

复制 RoundRobinLoadBalancer 内容 并修改 getInstanceResponse() 逻辑

package top.lingma.gateway.loadbalancer;

import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.springframework.beans.factory.ObjectProvider;
import org.springframework.cloud.client.ServiceInstance;
import org.springframework.cloud.client.loadbalancer.reactive.DefaultResponse;
import org.springframework.cloud.client.loadbalancer.reactive.EmptyResponse;
import org.springframework.cloud.client.loadbalancer.reactive.Request;
import org.springframework.cloud.client.loadbalancer.reactive.Response;
import org.springframework.cloud.loadbalancer.core.*;
import reactor.core.publisher.Mono;

import java.util.List;
import java.util.Random;
import java.util.concurrent.atomic.AtomicInteger;
import java.util.stream.Collectors;

public class AdvanceRoundRobinLoadBalancer implements ReactorServiceInstanceLoadBalancer {

    private static final Log log = LogFactory.getLog(AdvanceRoundRobinLoadBalancer.class);

    private final AtomicInteger position;
    private final AtomicInteger positionAdvance;

    @Deprecated
    private ObjectProvider<ServiceInstanceSupplier> serviceInstanceSupplier;

    private ObjectProvider<ServiceInstanceListSupplier> serviceInstanceListSupplierProvider;

    private final String serviceId;


    @Deprecated
    public AdvanceRoundRobinLoadBalancer(String serviceId, ObjectProvider<ServiceInstanceSupplier> serviceInstanceSupplier) {
        this(serviceId, serviceInstanceSupplier, new Random().nextInt(1000));
    }

    public AdvanceRoundRobinLoadBalancer(ObjectProvider<ServiceInstanceListSupplier> serviceInstanceListSupplierProvider, String serviceId) {
        this(serviceInstanceListSupplierProvider, serviceId, new Random().nextInt(1000));
    }


    public AdvanceRoundRobinLoadBalancer(ObjectProvider<ServiceInstanceListSupplier> serviceInstanceListSupplierProvider, String serviceId, int seedPosition) {
        this.serviceId = serviceId;
        this.serviceInstanceListSupplierProvider = serviceInstanceListSupplierProvider;
        this.position = new AtomicInteger(seedPosition);
        this.positionAdvance = new AtomicInteger(seedPosition);
    }

    @Deprecated
    public AdvanceRoundRobinLoadBalancer(String serviceId, ObjectProvider<ServiceInstanceSupplier> serviceInstanceSupplier, int seedPosition) {
        this.serviceId = serviceId;
        this.serviceInstanceSupplier = serviceInstanceSupplier;
        this.position = new AtomicInteger(seedPosition);
        this.positionAdvance = new AtomicInteger(seedPosition);
    }


    @Override

    public Mono<Response<ServiceInstance>> choose(Request request) {


        if (serviceInstanceListSupplierProvider != null) {
            ServiceInstanceListSupplier supplier = serviceInstanceListSupplierProvider.getIfAvailable(NoopServiceInstanceListSupplier::new);
            return supplier.get().next().map((instances) -> {
                // 此处做了选择逻辑的修改
                if (request instanceof AdvanceRequestContext) {
                    List<ServiceInstance> advanceInstance = instances.stream().filter(s -> s.getMetadata().getOrDefault("advance", "").equals("true")).collect(Collectors.toList());
                    return getInstanceResponse(advanceInstance, request);
                } else {
                    List<ServiceInstance> routineInstance = instances.stream().filter(s -> !s.getMetadata().getOrDefault("advance", "").equals("true")).collect(Collectors.toList());
                    return getInstanceResponse(routineInstance, request);
                }

            });
        }
        ServiceInstanceSupplier supplier = this.serviceInstanceSupplier.getIfAvailable(NoopServiceInstanceSupplier::new);
        return supplier.get().collectList().map((instances) -> {
            if (request instanceof AdvanceRequestContext) {
                // 此处做了选择逻辑的修改
                List<ServiceInstance> advanceInstance = instances.stream().filter(s -> s.getMetadata().getOrDefault("advance", "").equals("true")).collect(Collectors.toList());
                return getInstanceResponse(advanceInstance, request);
            } else {
                List<ServiceInstance> instance = instances.stream().filter(s -> !s.getMetadata().getOrDefault("advance", "").equals("true")).collect(Collectors.toList());
                return getInstanceResponse(instance, request);
            }

        });
    }

    private Response<ServiceInstance> getInstanceResponse(List<ServiceInstance> instances, Request request) {
        if (instances.isEmpty()) {
            if (request instanceof AdvanceRequestContext) {
                return new AdvanceEmptyResponse();
            }
            log.warn("No servers available for service: " + this.serviceId);
            return new EmptyResponse();
        }
        int pos = 1;
        //灰度发布选择逻辑
        if (request instanceof AdvanceRequestContext) {
            pos = Math.abs(this.positionAdvance.incrementAndGet());
        } else {
            pos = Math.abs(this.position.incrementAndGet());
        }
        ServiceInstance instance = instances.get(pos % instances.size());
        return new DefaultResponse(instance);

    }

}

AdvanceEmptyResponse 类是为了标识无灰度发布服务器，此时可以走正常服务器

package top.lingma.gateway.loadbalancer;

import org.springframework.cloud.client.ServiceInstance;
import org.springframework.cloud.client.loadbalancer.reactive.CompletionContext;
import org.springframework.cloud.client.loadbalancer.reactive.Response;

public class AdvanceEmptyResponse extends org.springframework.cloud.client.loadbalancer.EmptyResponse implements Response<ServiceInstance> {
    public AdvanceEmptyResponse() {
    }

    public void onComplete(CompletionContext completionContext) {
    }
}

AdvanceRequestContext 是为了能从 GlobalFilter 传递信息到 LoadBalancer

package top.lingma.gateway.loadbalancer;

import org.springframework.cloud.client.loadbalancer.reactive.Request;
import org.springframework.web.server.ServerWebExchange;

public class AdvanceRequestContext<T> implements Request {

    private T exchange;

    public AdvanceRequestContext(T exchange) {
        this.exchange = exchange;
    }

    @Override
    public T getContext() {
        return exchange;
    }
}

AdvanceReactiveLoadBalancerClientFilter 复制于 ReactiveLoadBalancerClientFilter

注意两点
第一灰度服务器选择在ReactiveLoadBalancerClientFilter 之前 LOAD_BALANCER_CLIENT_FILTER_ORDER = 10150 - 1;

package top.lingma.gateway.loadbalancer;

import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.springframework.cloud.client.ServiceInstance;
import org.springframework.cloud.client.loadbalancer.LoadBalancerUriTools;
import org.springframework.cloud.client.loadbalancer.reactive.Response;
import org.springframework.cloud.gateway.config.LoadBalancerProperties;
import org.springframework.cloud.gateway.filter.GatewayFilterChain;
import org.springframework.cloud.gateway.filter.GlobalFilter;
import org.springframework.cloud.gateway.filter.ReactiveLoadBalancerClientFilter;
import org.springframework.cloud.gateway.support.DelegatingServiceInstance;
import org.springframework.cloud.gateway.support.NotFoundException;
import org.springframework.cloud.loadbalancer.core.ReactorLoadBalancer;
import org.springframework.cloud.loadbalancer.core.ReactorServiceInstanceLoadBalancer;
import org.springframework.cloud.loadbalancer.support.LoadBalancerClientFactory;
import org.springframework.core.Ordered;
import org.springframework.stereotype.Component;
import org.springframework.web.server.ServerWebExchange;
import reactor.core.publisher.Mono;

import java.net.URI;
import java.util.List;

import static org.springframework.cloud.gateway.support.ServerWebExchangeUtils.*;

@Component
public class AdvanceReactiveLoadBalancerClientFilter implements GlobalFilter, Ordered {

    private static final Log log = LogFactory.getLog(ReactiveLoadBalancerClientFilter.class);

    private static final int LOAD_BALANCER_CLIENT_FILTER_ORDER = 10150 - 1;

    private final LoadBalancerClientFactory clientFactory;

    private LoadBalancerProperties properties;

    public AdvanceReactiveLoadBalancerClientFilter(LoadBalancerClientFactory clientFactory, LoadBalancerProperties properties) {
        this.clientFactory = clientFactory;
        this.properties = properties;
    }

    @Override
    public int getOrder() {
        return LOAD_BALANCER_CLIENT_FILTER_ORDER;
    }

    @Override
    @SuppressWarnings("Duplicates")
    public Mono<Void> filter(ServerWebExchange exchange, GatewayFilterChain chain) {
        // 灰度用户专属服务器         判定是否是灰度用户，是否拥有灰度权限 不然直接进行下一步
        List<String> secChUa = exchange.getRequest().getHeaders().get("sec-ch-ua");
        if (secChUa == null || secChUa.isEmpty() || !secChUa.stream().findFirst().map(r -> r.contains("Edge")).orElse(false)) {
            return chain.filter(exchange);
        }

        URI url = exchange.getAttribute(GATEWAY_REQUEST_URL_ATTR);
        String schemePrefix = exchange.getAttribute(GATEWAY_SCHEME_PREFIX_ATTR);
        if (url == null || (!"lb".equals(url.getScheme()) && !"lb".equals(schemePrefix))) {
            return chain.filter(exchange);
        }
        // preserve the original url
        addOriginalRequestUrl(exchange, url);

        if (log.isTraceEnabled()) {
            log.trace(ReactiveLoadBalancerClientFilter.class.getSimpleName() + " url before: " + url);
        }

        return choose(exchange).doOnNext(response -> {
            if (response instanceof AdvanceEmptyResponse) {
                return;
            }
            if (!response.hasServer()) {
                throw NotFoundException.create(properties.isUse404(), "Unable to find instance for " + url.getHost());
            }

            ServiceInstance retrievedInstance = response.getServer();

            URI uri = exchange.getRequest().getURI();

            // if the `lb:<scheme>` mechanism was used, use `<scheme>` as the default,
            // if the loadbalancer doesn't provide one.
            String overrideScheme = retrievedInstance.isSecure() ? "https" : "http";
            if (schemePrefix != null) {
                overrideScheme = url.getScheme();
            }

            DelegatingServiceInstance serviceInstance = new DelegatingServiceInstance(retrievedInstance, overrideScheme);

            URI requestUrl = reconstructURI(serviceInstance, uri);

            if (log.isTraceEnabled()) {
                log.trace("LoadBalancerClientFilter url chosen: " + requestUrl);
            }
            exchange.getAttributes().put(GATEWAY_REQUEST_URL_ATTR, requestUrl);
        }).then(chain.filter(exchange));
    }

    protected URI reconstructURI(ServiceInstance serviceInstance, URI original) {
        return LoadBalancerUriTools.reconstructURI(serviceInstance, original);
    }

    @SuppressWarnings("deprecation")
    private Mono<Response<ServiceInstance>> choose(ServerWebExchange exchange) {
        URI uri = exchange.getAttribute(GATEWAY_REQUEST_URL_ATTR);
        ReactorLoadBalancer<ServiceInstance> loadBalancer = this.clientFactory.getInstance(uri.getHost(), ReactorServiceInstanceLoadBalancer.class);
        if (loadBalancer == null) {
            throw new NotFoundException("No loadbalancer available for " + uri.getHost());
        }
        return loadBalancer.choose(createRequest(exchange));
    }

    /***
     * 此处进行了改造 传入了内容 方便后续 LoadBalancer 处理信息
     * @param exchange
     * @return
     */
    @SuppressWarnings("deprecation")
    private AdvanceRequestContext<ServerWebExchange> createRequest(ServerWebExchange exchange) {
        return new AdvanceRequestContext(exchange);
    }

}

以上已经完成了灰度发布的必要部分，再进行一下AutoConfiguration 注意，这里不能被Spring 扫描

package top.lingma.gateway.loadbalancer;

import org.springframework.boot.autoconfigure.condition.ConditionalOnMissingBean;
import org.springframework.cloud.client.ConditionalOnDiscoveryEnabled;
import org.springframework.cloud.client.ServiceInstance;
import org.springframework.cloud.loadbalancer.core.ReactorLoadBalancer;
import org.springframework.cloud.loadbalancer.core.ServiceInstanceListSupplier;
import org.springframework.cloud.loadbalancer.support.LoadBalancerClientFactory;
import org.springframework.context.annotation.Bean;
import org.springframework.core.env.Environment;

@ConditionalOnDiscoveryEnabled
public class AdvanceLoadBalancerAutoConfiguration {
    @Bean
    @ConditionalOnMissingBean
    public ReactorLoadBalancer<ServiceInstance> reactorServiceInstanceLoadBalancer(Environment environment, LoadBalancerClientFactory loadBalancerClientFactory) {
        String name = environment.getProperty(LoadBalancerClientFactory.PROPERTY_NAME);
        return new AdvanceRoundRobinLoadBalancer(loadBalancerClientFactory.getLazyProvider(name, ServiceInstanceListSupplier.class), name);
    }
}

最后 启动类配置 @LoadBalancerClients 的 defaultConfiguration

@SpringBootApplication()
@LoadBalancerClients(defaultConfiguration = AdvanceLoadBalancerAutoConfiguration.class)
public class LingmaGatewayApplication {

    public static void main(String[] args) {
        SpringApplication.run(LingmaGatewayApplication.class, args);
    }

}

关注公众号 [龗孖]