🔥🔥🔥提高服务器使用效率：合并微服务优化实践

前言

现在各个公司都在进行降本增效，其中的一个方面就是提高服务器的使用效率，由于在前期，错误的预估了业务的规模，导致了微服务拆分过细，很多服务器的使用在10%以下，为了解决这个情况，笔者根据eureka的注册原理，合并了多个服务，提高了服务的使用率。

目标

合并微服务势必会对前端的调用产生影响，所以此次我们的目的主要是降低对前端的影响。在不影响依赖方的情况下，将符合精简条件的server服务迁移至新服务中

调用流程目前各服务启动后各自注册自己的实例信息到注册中心，调用方通过feign客户端信息到注册中心拉取目标服务端的路由信息进行接口调用此处就不合大家讲具体的业务服务合并细节，我们先看一下eureka是如何支持一个服务注册多个服务名的。

原理分析

服务注册触发路径

java 复制代码

SpringApplication.run() -> 
this.refreshContext(context) -> 
this.refresh(context) -> 
ServletWebServerApplicationContext.refresh() -> 
this.finishRefresh() -> 
AbstractApplicationContext.finishRefresh -> 
DefaultLifecycleProcessor.onRefresh() -> 
this.startBeans -> 
this.start() -> 
this.doStart()

eureka的服务注册初始化加载是通过spring的SPI机制实现的

spring-cloud-starter-netflix-eureka-client -> spring.factories

java 复制代码

org.springframework.boot.autoconfigure.EnableAutoConfiguration=\
org.springframework.cloud.netflix.eureka.config.EurekaClientConfigServerAutoConfiguration,\
org.springframework.cloud.netflix.eureka.config.EurekaDiscoveryClientConfigServiceAutoConfiguration,\
org.springframework.cloud.netflix.eureka.EurekaClientAutoConfiguration,\
org.springframework.cloud.netflix.ribbon.eureka.RibbonEurekaAutoConfiguration,\
org.springframework.cloud.netflix.eureka.EurekaDiscoveryClientConfiguration

Eureka Client 的配置是通过 EurekaClientAutoConfiguration 和 EurekaDiscoveryClientConfiguration 进行处理的

java 复制代码

EurekaClientAutoConfiguration

public EurekaDiscoveryClient discoveryClient(EurekaClient client,
      EurekaClientConfig clientConfig) {
   return new EurekaDiscoveryClient(client, clientConfig);
}

eureka服务注册核心的类为 : EurekaServiceRegistry,通过这个类中提供的方法实现注册与下线，此外EurekaAutoServiceRegistration类实现了SmartLifecycle接口，会在spring容器加载和初始化完后调用start()方法

java 复制代码

public void start() {
		// only set the port if the nonSecurePort or securePort is 0 and this.port != 0
		if (this.port.get() != 0) {
			if (this.registration.getNonSecurePort() == 0) {
				this.registration.setNonSecurePort(this.port.get());
			}

			if (this.registration.getSecurePort() == 0 && this.registration.isSecure()) {
				this.registration.setSecurePort(this.port.get());
			}
		}

		// only initialize if nonSecurePort is greater than 0 and it isn't already running
		// because of containerPortInitializer below
		if (!this.running.get() && this.registration.getNonSecurePort() > 0) {

			this.serviceRegistry.register(this.registration);

			this.context.publishEvent(
					new InstanceRegisteredEvent<>(this, this.registration.getInstanceConfig()));
			this.runnin

eureka的注册信息全部保存在ApplicationInfoManager中

创建ApplicationInfoManager类的时候，创建了一个InstanceInfo类 ApplicationInfoManager通过EurekaInstanceConfigBean、InstanceInfo进行创建

java 复制代码

InstanceInfoReplicator

    public void run() {
        try {
            discoveryClient.refreshInstanceInfo();

            Long dirtyTimestamp = instanceInfo.isDirtyWithTime();
            if (dirtyTimestamp != null) {
                discoveryClient.register();
                instanceInfo.unsetIsDirty(dirtyTimestamp);
            }
        } catch (Throwable t) {
            logger.warn("There was a problem with the instance info replicator", t);
        } finally {
            Future next = scheduler.schedule(this, replicationIntervalSeconds, TimeUnit.SECONDS);
            scheduledPeriodicRef.set(next);
        }
    }

以上就是Eureka服务注册的主要流程，在最后附上配置多服务的配置类

java 复制代码

@Configuration
public class RegistryConfig implements SmartLifecycle {

    private static String appName = "server-name";
    public Map<String, CloudEurekaClient> applicationNameClientMap = new ConcurrentHashMap();
    @Autowired(required = false)
    private ObjectProvider<HealthCheckHandler> healthCheckHandler;
    @Autowired
    private InetUtils inetUtils;
    @Autowired
    private EurekaInstanceConfigBean eurekaInstanceConfigBean;
    @Autowired
    private EurekaClientConfig eurekaClientConfig;
    @Autowired
    private AbstractDiscoveryClientOptionalArgs<?> optionalArgs;
    @Autowired
    private ApplicationContext context;
    @Autowired
    private EurekaServiceRegistry serviceRegistr;
    private AtomicBoolean running = new AtomicBoolean(false);

    @Override
    public void start() {
        EurekaInstanceConfigBean eurekaInstanceConfigBean = new EurekaInstanceConfigBean(inetUtils);
        BeanUtils.copyProperties(this.eurekaInstanceConfigBean, eurekaInstanceConfigBean);
        eurekaInstanceConfigBean.setAppname(appName);
        eurekaInstanceConfigBean.setVirtualHostName(appName);
        eurekaInstanceConfigBean.setSecureVirtualHostName(appName);
        InstanceInfo instanceInfo = (new InstanceInfoFactory()).create(eurekaInstanceConfigBean);
        ApplicationInfoManager manager = new ApplicationInfoManager(eurekaInstanceConfigBean, instanceInfo);
        CloudEurekaClient cloudEurekaClient =
            new CloudEurekaClient(manager, eurekaClientConfig, this.optionalArgs, this.context);
        EurekaRegistration registration =
            EurekaRegistration.builder(eurekaInstanceConfigBean).with(manager).with(cloudEurekaClient)
                .with(this.healthCheckHandler).build();
        EurekaAutoServiceRegistration eurekaAutoServiceRegistration =
            new EurekaAutoServiceRegistration(this.context, this.serviceRegistr, registration);
        CloudEurekaClient mapResult = this.applicationNameClientMap.putIfAbsent(appName, cloudEurekaClient);
        if (mapResult == null) {
            eurekaAutoServiceRegistration.start();
        }
        this.running.set(true);
    }

    @Override
    public void stop() {
        CloudEurekaClient cloudEurekaClient = this.applicationNameClientMap.get(appName);
        if (cloudEurekaClient != null) {
            cloudEurekaClient.shutdown();
        }
        this.applicationNameClientMap.remove(appName);
        this.running.set(false);
    }

    @Override
    public boolean isRunning() {
        return this.running.get();
    }
}

尾声

eureka的技术栈已经有很多年了，很多公司已经在逐渐的淘汰，但其中还有很多设计的思想值得我们学习，在一开始，笔者也只是抱着试试看的心态来尝试一下，还好eureka的扩展足够支持，否则，只能让前端同事大量的修改调用的路径，测试同事进行大规模的回归测试了。大家要是有更好的方案也可以讨论一下，我的目的只是提供一些思路和引发讨论，以期能够共同进步。