十五.SpringCloud源码剖析-Ribbon工作流程

雨点打透心脏的1/2处 2022-12-11 15:03 170阅读 0赞

### 前言 ###

Ribbon是由Netflix公司开源的一个客户端负载均衡器，主要功能是实现服务之间的负载均衡调用，内置丰富的负载均衡算法，本章意在探讨Ribbon的核心工作流程，Ribbon基本使用请看《[SpringCloud极简入门-客户端负载均衡Ribbon][SpringCloud_-_Ribbon]》

### Ribbon的工作流程 ###

我们知道，微服务在启动成功之后，默认30s/次会从注册中心拉取服务注册表到本地缓存起来，而我们使用Ribbon时是通过RestTemplate发起请求，URL以：`http://user-server/user/...` 服务名方式去调用服务，其实Ribbon干的事情就是根据URL中的服务名去本地的服务注册表中查找服务名对应的服务实例(一个或多个)，然后通过负载均衡算法选择其中一个服务后，发起Http请求，那接下来我们就来看一下Ribbon的底层到底是怎么工作的

### Ribbon配合RestTemplate使用 ###

在《[SpringCloud极简入门-客户端负载均衡Ribbon][SpringCloud_-_Ribbon]》中我们有讲解到Ribbon的基本使用，首选需要定义RestTemplate

@SpringBootApplication
    @EnableEurekaClient
    public class OrderServerApplication1030
    { 
    
        //配置一个RestTemplate ，http客户端，支持Rest风格
        //@LoadBalanced :负载均衡注册，让RestTmplate可以实现负载均衡请求
        //这个标签标记RestTemplate可以使用LoadBalancerClient进行负载均衡
        @Bean
        @LoadBalanced
        public RestTemplate restTemplate(){ 
            return new RestTemplate();
        }
        //省略...

调用服务的是否使用服务名调用

@RestController
    public class OrderController { 
    
        //需要配置成Bean
        @Autowired
        private RestTemplate  restTemplate ;
    
        //浏览器调用该方法
        @RequestMapping(value = "/order/{id}",method = RequestMethod.GET)
        public User getById(@PathVariable("id")Long id){ 
            //发送http请求调用 user的服务，获取user对象 ： RestTemplate
            //user的ip,user的端口，user的Controller路径
            //String url = "http://localhost:1020/user/"+id;
            String url = "http://user-server/user/"+id;
    
            //发送http请求
            return restTemplate.getForObject(url, User.class);
    
        }
    }

被 @LoadBalanced标记的RestTemplate 可以使用LoadBalancerClient负载均衡客户端实现负载均衡，我们在使用RestTemplate 发起请求的时候需要跟上服务名的方式http://user-server/user/

### LoadBalancerClient负载均衡客户端 ###

我们从 @LoadBalanced入手，先看一下LoadBalancerClient它的源码

/** 注解标记RestTemplate 可以使用LoadBalancerClient 负载均衡客户端 * Annotation to mark a RestTemplate bean to be configured to use a LoadBalancerClient * @author Spencer Gibb */
    @Target({  ElementType.FIELD, ElementType.PARAMETER, ElementType.METHOD })
    @Retention(RetentionPolicy.RUNTIME)
    @Documented
    @Inherited
    @Qualifier
    public @interface LoadBalanced { 
    }

这个注解@LoadBalanced 的作用在注释上说的非常清楚，就是标记RestTemplate课程可以使用使用LoadBalancerClient来实现负载均衡，LoadBalancerClient就是Ribbon实现负载均衡的一个客户端，它在spring-cloud-commons包下，我们可以直接看LoadBalancerClient的源码

/** 客户端负载均衡 * Represents a client side load balancer * @author Spencer Gibb */
    public interface LoadBalancerClient extends ServiceInstanceChooser { 
    	//执行请求，会根据serviceId使用负载均衡查找服务
    	/** 使用负载均衡器执行指定服务 * execute request using a ServiceInstance from the LoadBalancer for the specified service * 服务Id - 服务ID来查找负载平衡器 * @param serviceId the service id to look up the LoadBalancer * * @param request allows implementations to execute pre and post actions such as * incrementing metrics * 返回选择的服务 * @return the result of the LoadBalancerRequest callback on the selected * ServiceInstance */
    	<T> T execute(String serviceId, LoadBalancerRequest<T> request) throws IOException;
    
    	//执行请求，这个方法多了一个参数ServiceInstance ，即请求指定的服务
    	/** * execute request using a ServiceInstance from the LoadBalancer for the specified * service * @param serviceId the service id to look up the LoadBalancer * @param serviceInstance the service to execute the request to * @param request allows implementations to execute pre and post actions such as * incrementing metrics * @return the result of the LoadBalancerRequest callback on the selected * ServiceInstance */
    	<T> T execute(String serviceId, ServiceInstance serviceInstance, LoadBalancerRequest<T> request) throws IOException;
    
    	//重构URL，把http://myservice/path/to/service重构成http://ip:端口/path/to/service
    	/** * Create a proper URI with a real host and port for systems to utilize. * Some systems use a URI with the logical serivce name as the host, * such as http://myservice/path/to/service. This will replace the * service name with the host:port from the ServiceInstance. * @param instance * @param original a URI with the host as a logical service name * @return a reconstructed URI */
    	URI reconstructURI(ServiceInstance instance, URI original);
    }

LoadBalancerClient接口三个方法，excute()为执行请求，reconstructURI()用来重构url，它实现了ServiceInstanceChooser 接口，这个接口的作用是用来选择服务的，看下源码

/** 通过使用负载平衡器，选择一个服务器发送请求。 * Implemented by classes which use a load balancer to choose a server to * send a request to. * * @author Ryan Baxter */
    public interface ServiceInstanceChooser { 
    
        /** 从LoadBalancer中为指定服务选择一个ServiceInstance * Choose a ServiceInstance from the LoadBalancer for the specified service * * //根据服务id去LoadBalancer查找服务 * @param serviceId the service id to look up the LoadBalancer * * 返回查找到的服务实例ServiceInstance * @return a ServiceInstance that matches the serviceId */
        ServiceInstance choose(String serviceId);
    }

提供了一个choose方法，根据服务ID serviceId 查找一个ServiceInstance 服务实例，这里的serviceId其实就是http://user-server/… url中带的服务名。

LoadBalancerClient 还有一个默认实现类RibbonLoadBalancerClient，这个实现是针对Ribbon的客户端负载均衡，继承关系如下：  
![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3UwMTQ0OTQxNDg_size_16_color_FFFFFF_t_70_pic_center]

RibbonLoadBalancerClient是一个非常核心的类，最终的负载均衡的请求处理由它来执行，源码如下：

public class RibbonLoadBalancerClient implements LoadBalancerClient { 
    
    	private SpringClientFactory clientFactory;
    
    	public RibbonLoadBalancerClient(SpringClientFactory clientFactory) { 
    		this.clientFactory = clientFactory;
    	}
    	//重构URL，找到服务之后，把http://服务名/ 格式 重构成 http://ip:port/ 格式
    	@Override
    	public URI reconstructURI(ServiceInstance instance, URI original) { 
    		Assert.notNull(instance, "instance can not be null");
    		String serviceId = instance.getServiceId();
    		RibbonLoadBalancerContext context = this.clientFactory
    				.getLoadBalancerContext(serviceId);
    
    		URI uri;
    		Server server;
    		if (instance instanceof RibbonServer) { 
    			RibbonServer ribbonServer = (RibbonServer) instance;
    			server = ribbonServer.getServer();
    			uri = updateToSecureConnectionIfNeeded(original, ribbonServer);
    		} else { 
    			server = new Server(instance.getScheme(), instance.getHost(), instance.getPort());
    			IClientConfig clientConfig = clientFactory.getClientConfig(serviceId);
    			ServerIntrospector serverIntrospector = serverIntrospector(serviceId);
    			uri = updateToSecureConnectionIfNeeded(original, clientConfig,
    					serverIntrospector, server);
    		}
    		return context.reconstructURIWithServer(server, uri);
    	}
    	//根据服务名，查找服务实例，选择一个返回ServiceInstance 
    	@Override
    	public ServiceInstance choose(String serviceId) { 
    		//查找服务
    		Server server = getServer(serviceId);
    		if (server == null) { 
    			return null;
    		}
    		return new RibbonServer(serviceId, server, isSecure(server, serviceId),
    				serverIntrospector(serviceId).getMetadata(server));
    	}
    	//执行请求
    	@Override
    	public <T> T execute(String serviceId, LoadBalancerRequest<T> request) throws IOException { 
    		//获取负载均衡器[重要]
    		ILoadBalancer loadBalancer = getLoadBalancer(serviceId);
    		//选择服务，使用负载均衡器，根据服务的ID，选择一个服务
    		Server server = getServer(loadBalancer);
    		if (server == null) { 
    			throw new IllegalStateException("No instances available for " + serviceId);
    		}
    		//选择的服务封装成一个RibbonServer:RibbonServer implements ServiceInstance
    		RibbonServer ribbonServer = new RibbonServer(serviceId, server, isSecure(server,
    				serviceId), serverIntrospector(serviceId).getMetadata(server));
    		//执行请求调用服务
    		return execute(serviceId, ribbonServer, request);
    	}
    	//执行请求调用服务
    	@Override
    	public <T> T execute(String serviceId, ServiceInstance serviceInstance, LoadBalancerRequest<T> request) throws IOException { 
    		Server server = null;
    		if(serviceInstance instanceof RibbonServer) { 
    			server = ((RibbonServer)serviceInstance).getServer();
    		}
    		if (server == null) { 
    			throw new IllegalStateException("No instances available for " + serviceId);
    		}
    
    		RibbonLoadBalancerContext context = this.clientFactory
    				.getLoadBalancerContext(serviceId);
    		RibbonStatsRecorder statsRecorder = new RibbonStatsRecorder(context, server);
    
    		try { 
    			//使用 LoadBalancerRequest 向服务发请求
    			T returnVal = request.apply(serviceInstance);
    			statsRecorder.recordStats(returnVal);
    			return returnVal;
    		}
    		// catch IOException and rethrow so RestTemplate behaves correctly
    		catch (IOException ex) { 
    			statsRecorder.recordStats(ex);
    			throw ex;
    		}
    		catch (Exception ex) { 
    			statsRecorder.recordStats(ex);
    			ReflectionUtils.rethrowRuntimeException(ex);
    		}
    		return null;
    	}
    
    	private ServerIntrospector serverIntrospector(String serviceId) { 
    		ServerIntrospector serverIntrospector = this.clientFactory.getInstance(serviceId,
    				ServerIntrospector.class);
    		if (serverIntrospector == null) { 
    			serverIntrospector = new DefaultServerIntrospector();
    		}
    		return serverIntrospector;
    	}
    	//是否是https请求
    	private boolean isSecure(Server server, String serviceId) { 
    		IClientConfig config = this.clientFactory.getClientConfig(serviceId);
    		ServerIntrospector serverIntrospector = serverIntrospector(serviceId);
    		return RibbonUtils.isSecure(config, serverIntrospector, server);
    	}
    	//根据服务ID选择服务
    	protected Server getServer(String serviceId) { 
    		return getServer(getLoadBalancer(serviceId));
    	}
    	
    	//负载均衡器选择服务
    	protected Server getServer(ILoadBalancer loadBalancer) { 
    		if (loadBalancer == null) { 
    			return null;
    		}
    		return loadBalancer.chooseServer("default"); // TODO: better handling of key
    	}
    	//根据服务id得到负载均衡器
    	protected ILoadBalancer getLoadBalancer(String serviceId) { 
    		return this.clientFactory.getLoadBalancer(serviceId);
    	}
    ...省略...

解释一下：

*  这里的`ServiceInstance choose(String serviceId)`方法的作用是根据ServideId选择一个服务，底层实现是通过`LoadBalancer.chooseServer` 负载均衡器LoadBalancer来完成的服务的选择的
 *  选择到服务之后调用`execute`向选择到的服务发起请求，通过LoadBalancerRequest来完成其请求。

### RestTemplate的执行流程 ###

RestTmplate发请求时地址 `"http://user-server/user/"+id`; 中 user-server是当前服务需要调用的目标服务的服务名，那么Ribbon到底是如何实现负载均衡调用的呢？我们可以从这里跟踪一下RestTemplate的执行流程

public class RestTemplate extends InterceptingHttpAccessor implements RestOperations { 
    ...省略...
     @Nullable
        protected <T> T doExecute(URI url, @Nullable HttpMethod method, @Nullable RequestCallback requestCallback, @Nullable ResponseExtractor<T> responseExtractor) throws RestClientException { 
            Assert.notNull(url, "URI is required");
            Assert.notNull(method, "HttpMethod is required");
            ClientHttpResponse response = null;
    
            Object var14;
            try { 
            	//创建请求对象，使用SimpleClientHttpRequestFactory创建ClientHttpRequest 
                ClientHttpRequest request = this.createRequest(url, method);
                if (requestCallback != null) { 
                	//设置header和body
                    requestCallback.doWithRequest(request);
                }
    
                response = request.execute();
                this.handleResponse(url, method, response);
                var14 = responseExtractor != null ? responseExtractor.extractData(response) : null;
            } catch (IOException var12) { 
                String resource = url.toString();
                String query = url.getRawQuery();
                resource = query != null ? resource.substring(0, resource.indexOf(63)) : resource;
                throw new ResourceAccessException("I/O error on " + method.name() + " request for \"" + resource + "\": " + var12.getMessage(), var12);
            } finally { 
                if (response != null) { 
                    response.close();
                }
    
            }
    
            return var14;
        }

请求来到`RestTemplate#doExecute`方法，首选是通过使用`SimpleClientHttpRequestFactory`根据url和method创建`ClientHttpRequest` 请求对象,使用的实现是`InterceptingClientHttpRequestFactory`，然后使用`response = request.execute();`去执行请求,一路跟踪，请求来到InterceptingClientHttpRequest\#executeInternal

class InterceptingClientHttpRequest extends AbstractBufferingClientHttpRequest { 
    	//headers请求头 ， bufferedOutput输出内容
        protected final ClientHttpResponse executeInternal(HttpHeaders headers, byte[] bufferedOutput) throws IOException { 
        	//创建拦截器执行器
            InterceptingClientHttpRequest.InterceptingRequestExecution requestExecution = new InterceptingClientHttpRequest.InterceptingRequestExecution();
            return requestExecution.execute(this, bufferedOutput);
        }

这里通过`InterceptingClientHttpRequest.InterceptingRequestExecution()` 拦截器执行器去执行请求,请求来到InterceptingClientHttpRequest.InterceptingRequestExecution\#execute

private class InterceptingRequestExecution implements ClientHttpRequestExecution { 
            private final Iterator<ClientHttpRequestInterceptor> iterator;
    
            public InterceptingRequestExecution() { 
                this.iterator = InterceptingClientHttpRequest.this.interceptors.iterator();
            }
    
            public ClientHttpResponse execute(HttpRequest request, byte[] body) throws IOException { 
                if (this.iterator.hasNext()) { 
                	//[重要]这里取到的正是 LoadBalancerInterceptor
                    ClientHttpRequestInterceptor nextInterceptor = (ClientHttpRequestInterceptor)this.iterator.next();
                    return nextInterceptor.intercept(request, body, this);
                } else { 
                    HttpMethod method = request.getMethod();
                    Assert.state(method != null, "No standard HTTP method");
                    //如果iterator中没有拦截器了，就创建一个ClientHttpRequest去执行请求
                    ClientHttpRequest delegate = InterceptingClientHttpRequest.this.requestFactory.createRequest(request.getURI(), method);
                    request.getHeaders().forEach((key, value) -> { 
                        delegate.getHeaders().addAll(key, value);
                    });
                    if (body.length > 0) { 
                        if (delegate instanceof StreamingHttpOutputMessage) { 
                            StreamingHttpOutputMessage streamingOutputMessage = (StreamingHttpOutputMessage)delegate;
                            streamingOutputMessage.setBody((outputStream) -> { 
                                StreamUtils.copy(body, outputStream);
                            });
                        } else { 
                            StreamUtils.copy(body, delegate.getBody());
                        }
                    }
    				//执行请求
                    return delegate.execute();
                }
            }
        }

`InterceptingRequestExecution` 中维护了一个`Iterator<ClientHttpRequestInterceptor>` iterator;其中`LoadBalancerInterceptor` 就在该集合中，所以请求来到`LoadBalancerInterceptor #intercept(request, body, this);` 方法

//负载均衡拦截器
    public class LoadBalancerInterceptor implements ClientHttpRequestInterceptor { 
    	//负载均衡客户端[重要]
    	private LoadBalancerClient loadBalancer;
    	//负载均衡请求创建工厂
    	private LoadBalancerRequestFactory requestFactory;
    	//初始化
    	public LoadBalancerInterceptor(LoadBalancerClient loadBalancer, LoadBalancerRequestFactory requestFactory) { 
    		this.loadBalancer = loadBalancer;
    		this.requestFactory = requestFactory;
    	}
    	//初始化
    	public LoadBalancerInterceptor(LoadBalancerClient loadBalancer) { 
    		// for backwards compatibility
    		this(loadBalancer, new LoadBalancerRequestFactory(loadBalancer));
    	}
    	//拦截器核心方法【重要】
    	//request请求对象
    	//body 内容
    	@Override
    	public ClientHttpResponse intercept(final HttpRequest request, final byte[] body,
    			final ClientHttpRequestExecution execution) throws IOException { 
    		//请求的URL，格式如：http://user-server/user/1 ，user-server是服务名
    		final URI originalUri = request.getURI();
    		//URL中的服务名
    		String serviceName = originalUri.getHost();
    		
    		Assert.state(serviceName != null, "Request URI does not contain a valid hostname: " + originalUri);
    		//通过requestFactory.createRequest(request, body, execution)创建LoadBalancerRequest
    		//然后调用负载均衡器执行请求，参数：服务名，LoadBalancerRequest
    		return this.loadBalancer.execute(serviceName, requestFactory.createRequest(request, body, execution));
    	}
    }

这里蛮重要的，请求调用了`LoadBalancerInterceptor #intercept`负载均衡拦截器的拦截方法，获取到URL，从中获取到主机名即调用的服务名(Ribbon客户端服务名)，然后使用`LoadBalancerRequestFactory` 创建了`LoadBalancerRequest`请求对象，调用`loadBalancer#execute` 负载均衡器执行请求

### ILoadBalancer 选择服务(负载均衡) ###

请求来到`RibbonLoadBalancerClient#execute`

@Override
    	public <T> T execute(String serviceId, LoadBalancerRequest<T> request) throws IOException { 
    		//获取负载均衡器
    		ILoadBalancer loadBalancer = getLoadBalancer(serviceId);
    		//loadBalancer选择服务
    		Server server = getServer(loadBalancer);
    		if (server == null) { 
    			throw new IllegalStateException("No instances available for " + serviceId);
    		}
    		//选择的服务封装成RibbonServer 
    		RibbonServer ribbonServer = new RibbonServer(serviceId, server, isSecure(server,
    				serviceId), serverIntrospector(serviceId).getMetadata(server));
    		//LoadBalancerRequest对服务执行请求
    		return execute(serviceId, ribbonServer, request);
    	}

这里就蛮关键了

*  首选是通过服务名调用getLoadBalancer方法得到负载均衡器
 *  然后getServer(loadBalancer)是通过负载均衡器选择一个服务，底层会使用IRule的算法
 *  然后将服务封装成RibbonServer 对象，交给LoadBalancerRequest去执行请求

这里的负载均衡器默认会走ZoneAwareLoadBalancer，它是通过SpringClientFactory 从Ribbon上下文对象中获取到的负载均衡器对象,关于这个我们在上一章讨论过

public class RibbonLoadBalancerClient implements LoadBalancerClient { 
    ...省略...
    	private SpringClientFactory clientFactory;
    	protected ILoadBalancer getLoadBalancer(String serviceId) { 
    		return this.clientFactory.getLoadBalancer(serviceId);
    	}

而得到ILoadBalancer之后，调用getServer(loadBalancer)方法选择服务，我们跟踪一下

public class RibbonLoadBalancerClient implements LoadBalancerClient { 
    ...省略...
    	protected Server getServer(ILoadBalancer loadBalancer) { 
    		if (loadBalancer == null) { 
    			return null;
    		}
    		//ZoneAwareLoadBalancer#chooseServer
    		return loadBalancer.chooseServer("default"); // TODO: better handling of key
    	}

这里`loadBalancer.chooseServer("default");`请求来到`ZoneAwareLoadBalancer#chooseServer`,源码如下：

public class ZoneAwareLoadBalancer<T extends Server> extends DynamicServerListLoadBalancer<T> { 
    ...省略...
    @Override
        public Server chooseServer(Object key) { 
            if (!ENABLED.get() || getLoadBalancerStats().getAvailableZones().size() <= 1) { 
            	//如果禁用了zone,或者自由一个zone会走这里
                logger.debug("Zone aware logic disabled or there is only one zone");
                return super.chooseServer(key);
            }
    		//下面就是根据zone选择服务了，默认情况下不会走下面
            Server server = null;
            try { 
                LoadBalancerStats lbStats = getLoadBalancerStats();
                //得到zone快照
                Map<String, ZoneSnapshot> zoneSnapshot = ZoneAvoidanceRule.createSnapshot(lbStats);
                logger.debug("Zone snapshots: {}", zoneSnapshot);
                if (triggeringLoad == null) { 
                    triggeringLoad = DynamicPropertyFactory.getInstance().getDoubleProperty(
                            "ZoneAwareNIWSDiscoveryLoadBalancer." + this.getName() + ".triggeringLoadPerServerThreshold", 0.2d);
                }
    
                if (triggeringBlackoutPercentage == null) { 
                    triggeringBlackoutPercentage = DynamicPropertyFactory.getInstance().getDoubleProperty(
                            "ZoneAwareNIWSDiscoveryLoadBalancer." + this.getName() + ".avoidZoneWithBlackoutPercetage", 0.99999d);
                }
                //得到可用的zone
                Set<String> availableZones = ZoneAvoidanceRule.getAvailableZones(zoneSnapshot, triggeringLoad.get(), triggeringBlackoutPercentage.get());
                logger.debug("Available zones: {}", availableZones);
                if (availableZones != null &&  availableZones.size() < zoneSnapshot.keySet().size()) { 
                	//随机选择区域
                    String zone = ZoneAvoidanceRule.randomChooseZone(zoneSnapshot, availableZones);
                    logger.debug("Zone chosen: {}", zone);
                    if (zone != null) { 
                        BaseLoadBalancer zoneLoadBalancer = getLoadBalancer(zone);
                        //选择服务
                        server = zoneLoadBalancer.chooseServer(key);
                    }
                }
            } catch (Exception e) { 
                logger.error("Error choosing server using zone aware logic for load balancer={}", name, e);
            }
            if (server != null) { 
                return server;
            } else { 
                logger.debug("Zone avoidance logic is not invoked.");
                return super.chooseServer(key);
            }
        }

这里做了一个判断，如果没有设置zone或者只有一个zone(默认)，这里会调用 `return super.chooseServer(key);`通过父类的`BaseLoadBalancer#chooseServer`方法选择服务，这也是默认的执行流程，代码走到了`BaseLoadBalancer#chooseServer`方法中，源码如下

public class BaseLoadBalancer extends AbstractLoadBalancer implements
            PrimeConnections.PrimeConnectionListener, IClientConfigAware { 
     public Server chooseServer(Object key) { 
            if (counter == null) { 
            	//创建一个计数器
                counter = createCounter();
            }
            //计数器增加
            counter.increment();
            //如果负载均衡规则为空，返回空
            if (rule == null) { 
                return null;
            } else { 
                try { 
                	//[重要]调用了负载均衡器算法类的choose方法
                    return rule.choose(key);
                } catch (Exception e) { 
                    logger.warn("LoadBalancer [{}]: Error choosing server for key {}", name, key, e);
                    return null;
                }
            }
        }

在`BaseLoadBalancer #chooseServer`方法中调用了`IRule#choose`方法进行服务的选择服务，IRule有很多是算法策略实现类，默认会走轮询算法，如果有定义负载均衡算法，这里rule.choose调用的就是定义的算法类

这里我打了个端点，跟踪了一下源码发现默认情况下会从`BaseLoadBalancer#chooseServer`方法中调用`PredicateBasedRule#choose` ，PredicateBasedRule本身是继承`ClientConfigEnabledRoundRobinRule`，也就是说`PredicateBasedRule`是使用的是轮询算法，同时它扩展了Predicate功能，即：提供了服务器过滤逻辑

/** 一个规则，提供了服务器过滤逻辑，具体使用的是AbstractServerPredicate实现过滤功能。 过滤后，服务器从过滤列表中的循环方式返回。 * A rule which delegates the server filtering logic to an instance of {@link AbstractServerPredicate}. * After filtering, a server is returned from filtered list in a round robin fashion. * * * @author awang * */
    public abstract class PredicateBasedRule extends ClientConfigEnabledRoundRobinRule { 
       
        /** 抽象函数，返回AbstractServerPredicate，用来对服务做过滤的 * Method that provides an instance of {@link AbstractServerPredicate} to be used by this class. * */
        public abstract AbstractServerPredicate getPredicate();
            
        /** * Get a server by calling {@link AbstractServerPredicate#chooseRandomlyAfterFiltering(java.util.List, Object)}. * The performance for this method is O(n) where n is number of servers to be filtered. */
        @Override
        public Server choose(Object key) { 
        	//得到负载均衡器
            ILoadBalancer lb = getLoadBalancer();
             //通过AbstractServerPredicate的chooseRoundRobinAfterFiltering选出具体的服务实例
            //AbstractServerPredicate的子类实现的Predicate逻辑来过滤一部分服务实例
            //然后在以线性轮询的方式从过滤后的实例中选出一个
            Optional<Server> server = getPredicate().chooseRoundRobinAfterFiltering(lb.getAllServers(), key);
            if (server.isPresent()) { 
                return server.get();
            } else { 
                return null;
            }       
        }
    }

这里使用了`AbstractServerPredicate#chooseRoundRobinAfterFiltering`来选择服务从lb.getAllServers()得到所有的服务作为参数，继续跟踪下去

/** * Choose a server in a round robin fashion after the predicate filters a given list of servers and load balancer key. */
        public Optional<Server> chooseRoundRobinAfterFiltering(List<Server> servers, Object loadBalancerKey) { 
        	//得到合格的服务列表，主要根据zone做一个过滤
            List<Server> eligible = getEligibleServers(servers, loadBalancerKey);
            if (eligible.size() == 0) { 
              //没找到合格的服务
                return Optional.absent();
            }
            //以线性轮询的方式合格的服务列表获取一个实例
            //incrementAndGetModulo方法会以轮询的方式计算一个下标值
            return Optional.of(eligible.get(incrementAndGetModulo(eligible.size())));
        }
    	...省略...
     /** 引用于 RoundRobinRule 算法策略 ， 轮询 * Referenced from RoundRobinRule * Inspired by the implementation of {@link AtomicInteger#incrementAndGet()}. * * @param modulo The modulo to bound the value of the counter. * @return The next value. */
         //增量和取模实现轮询
        private int incrementAndGetModulo(int modulo) { 
            for (;;) { 
                int current = nextIndex.get();
                int next = (current + 1) % modulo;
                if (nextIndex.compareAndSet(current, next) && current < modulo)
                    return current;
            }
        }

这里首先会通过zone过滤出可用的服务列表，然后使用轮询算法选择一个服务返回，到这里选择服务的流程调用

### LoadBalancerRequest 执行服务 ###

代码继续回到 RibbonLoadBalancerClient\#execute，选择完服务之后，服务被封装成RibbonServer

@Override
    	public <T> T execute(String serviceId, LoadBalancerRequest<T> request) throws IOException { 
    		//得到负载均衡器
    		ILoadBalancer loadBalancer = getLoadBalancer(serviceId);
    		//选择服务
    		Server server = getServer(loadBalancer);
    		if (server == null) { 
    			throw new IllegalStateException("No instances available for " + serviceId);
    		}
    		//把server 封装成RibbonServer 
    		RibbonServer ribbonServer = new RibbonServer(serviceId, server, isSecure(server,
    				serviceId), serverIntrospector(serviceId).getMetadata(server));
    		//执行服务调用
    		return execute(serviceId, ribbonServer, request);
    	}

找到服务后，调用了execute方法执行后续请求

@Override
    	public <T> T execute(String serviceId, ServiceInstance serviceInstance, LoadBalancerRequest<T> request) throws IOException { 
    		Server server = null;
    		if(serviceInstance instanceof RibbonServer) { 
    			server = ((RibbonServer)serviceInstance).getServer();
    		}
    		if (server == null) { 
    			throw new IllegalStateException("No instances available for " + serviceId);
    		}
    		//加载Ribbon负载均衡器上下文对象
    		RibbonLoadBalancerContext context = this.clientFactory
    				.getLoadBalancerContext(serviceId);
    		RibbonStatsRecorder statsRecorder = new RibbonStatsRecorder(context, server);
    
    		try { 
    			//LoadBalancerRequest.apply执行请求
    			T returnVal = request.apply(serviceInstance);
    			statsRecorder.recordStats(returnVal);
    			return returnVal;
    		}
    		// catch IOException and rethrow so RestTemplate behaves correctly
    		catch (IOException ex) { 
    			statsRecorder.recordStats(ex);
    			throw ex;
    		}
    		catch (Exception ex) { 
    			statsRecorder.recordStats(ex);
    			ReflectionUtils.rethrowRuntimeException(ex);
    		}
    		return null;
    	}

这里调用`LoadBalancerRequest.apply`执行请求，后面还会调用`LoadBalancerRequestFactory#createRequest`方法创建请求，调用`ClientHttpRequestExecution#execute`执行，然后又会将请求委派给`ClientHttpRequest#execute`去执行，再往后面走就是创建`HttpURLConnection`链接对象发送请求了，我们就不继续跟下去了。

### 总结 ###

纵观Ribbon的工作流程大致如下

*  初始化的时候创建好Ribbon的上下文，以及相关的组件，如ILoadBalancer，IConfig,IRule等等
 *  初始化过程中会通过ServerList从EurekaClient加载负载均衡候选的服务列表，并定时更新服务列表，使用ServerListFilter过滤之后，使用IPing检查是否更新服务列表
 *  被注解了@LoadBalance标签的RestTemplate可以使用LoadBalancerClient作负载均衡，并添加好拦截器LoadBalancerInterceptor
 *  当请求发起RestTemplate会把请求交给LoadBalancerInterceptor 拦截器，LoadBalancerInterceptor 拦截器调用
 *  LoadBalancerClient接收到请求使用ILoadBalancer负载均衡器选择服务，底层用到IRule算法
 *  选择好服务之后，LoadBalancerClient把请求交给LoadBalancerRequest去执行  
    ![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3UwMTQ0OTQxNDg_size_16_color_FFFFFF_t_70_pic_center 1]

[SpringCloud_-_Ribbon]: https://blog.csdn.net/u014494148/article/details/105002095
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3UwMTQ0OTQxNDg_size_16_color_FFFFFF_t_70_pic_center]: /images/20221123/a82e40dff23e4c83994a4c915c82c7b1.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3UwMTQ0OTQxNDg_size_16_color_FFFFFF_t_70_pic_center 1]: /images/20221123/e9fd13f367774ec098eeef82c505af05.png