导读
- 上一篇我们分析了 Dubbo设计之ExtensionLoader,了解了其原理之后,就方便我们分析Dubbo客户端得初始化过程了。
- 关键字 :Dubbo客户端初始化、Netty入门级用法
- Dubbo版本 : 2.7+
- 入口: ReferenceConfig
客户端初始化:
- 下面是Dubbo消费者API使用方式的示例代码
// 创建ReferenceConfig,此类会有一系列的初始化准备过程
ReferenceConfig reference = new ReferenceConfig<>();
// 设置消费者应用配置
reference.setApplication(new ApplicationConfig("dubbo-demo-api-consumer"));
// 设置注册地址
reference.setRegistry(new RegistryConfig("zookeeper://" + zookeeperHost + ":" + zookeeperPort));
// 设置引用的API接口
reference.setInterface(AlipayInfoAbility.class);
// 关闭服务启动检查(建议不要关闭,找不到服务要在应用发布时就要解决)
reference.setCheck(false);
// 获取引用
AlipayInfoAbility service = reference.get();
上述代码中, reference.get()
是整个调用方获取接口代理得核心起始点。主要核心逻辑如下:
① 设置接口级别得公共参数 Mapref
做准备,例如:设置key = side,value = consumer,表示当前是服务调用方;通过 MetricsConfig
、RegistryConfig(注册中心)
、ConfigCenterConfig(配置中心)
、ApplicationConfig
、ModuleConfig
、ConsumerConfig(可以指定消费端线程池类型:cached, fixed, limit, eager;核心线程数、最大线程数、队列大小、共享连接数shareconnections(TCP长连接得客户端数量))
、MethodConfig(可以指定方法相关得参数配置:可以设置重试次数、并发数(executes)、是否启动session粘连(sticky)、方法是否有异步回调等)
等初始化配置来填充 map。
② 接着就是调用创建代理引用得核心方法org.apache.dubbo.config.ReferenceConfig#createProxy
private T createProxy(Map map) {
if (shouldJvmRefer(map)) { // 1.判断是否是本地调用(通过scope=local、injvm=false都可以),重要用于本地测试,防止调用远程服务器
URL url = new URL(LOCAL_PROTOCOL, LOCALHOST_VALUE, 0, interfaceClass.getName()).addParameters(map);
invoker = REF_PROTOCOL.refer(interfaceClass, url); // 通过自适应扩展点机制获取代理Invoker
if (logger.isInfoEnabled()) {
logger.info("Using injvm service " + interfaceClass.getName());
}
} else { // 2.不是本地调用
urls.clear(); // reference retry init will add url to urls, lead to OOM
if (url != null && url.length() > 0) { // user specified URL, could be peer-to-peer address, or register center's address.
String[] us = SEMICOLON_SPLIT_PATTERN.split(url);
if (us != null && us.length > 0) {
for (String u : us) {
URL url = URL.valueOf(u);
if (StringUtils.isEmpty(url.getPath())) {
url = url.setPath(interfaceName);
}
if (REGISTRY_PROTOCOL.equals(url.getProtocol())) { // 3.协议类型是注册(registry)
urls.add(url.addParameterAndEncoded(REFER_KEY, StringUtils.toQueryString(map)));
} else {
urls.add(ClusterUtils.mergeUrl(url, map));
}
}
}
} else { // assemble URL from register center's configuration
// if protocols not injvm checkRegistry
if (!LOCAL_PROTOCOL.equalsIgnoreCase(getProtocol())){ // 若注册协议不是injvm
// 校验注册中心配置(RegistryConfig),会向后兼容-Ddubbo.registry.address启动参数
checkRegistry();
// 4.通过配置中心相关设置解析成URL(一般注册中心都使用 zookeeper,没配置默认是dubbo)
// 执行完之后会将protocol属性key设置为registry,当下面调用协议自适应扩展点得时候就会找到 Registryprotocol
List us = loadRegistries(false);
if (CollectionUtils.isNotEmpty(us)) {
for (URL u : us) {
// 5.加载监控配置,显式设置直接使用。没显式设置:判断是否有logstat监控扩展点工厂,没有会使用Dubbo得监控
URL monitorUrl = loadMonitor(u);
if (monitorUrl != null) {
map.put(MONITOR_KEY, URL.encode(monitorUrl.toFullString()));
}
urls.add(u.addParameterAndEncoded(REFER_KEY, StringUtils.toQueryString(map)));
}
}
if (urls.isEmpty()) {
throw new IllegalStateException("No such any registry to reference " + interfaceName + " on the consumer " + NetUtils.getLocalHost() + " use dubbo version " + Version.getVersion() + ", please config to your spring config.");
}
}
}
if (urls.size() == 1) {// 注册中心单节点
// 6.通过Registryprotocol$Apaptive 自适应扩展点实现获取Invoker
invoker = REF_PROTOCOL.refer(interfaceClass, urls.get(0));
} else { // 注册中心集群
List> invokers = new ArrayList>();
URL registryURL = null;
for (URL url : urls) {
// 7.通过Registryprotocol$Apaptive获取Invoker,放入到列表中
invokers.add(REF_PROTOCOL.refer(interfaceClass, url));
if (REGISTRY_PROTOCOL.equals(url.getProtocol())) {
registryURL = url; // use last registry url
}
}
if (registryURL != null) { // registry url is available
// use RegistryAwareCluster only when register's CLUSTER is available
URL u = registryURL.addParameter(CLUSTER_KEY, RegistryAwareCluster.NAME);
// The invoker wrap relation would be: RegistryAwareClusterInvoker(StaticDirectory) -> FailoverClusterInvoker(RegistryDirectory, will execute route) -> Invoker
invoker = CLUSTER.join(new StaticDirectory(u, invokers));// 8.通过集群RegistryAwareCluster对Invoker进行包装
} else { // not a registry url, must be direct invoke.
invoker = CLUSTER.join(new StaticDirectory(invokers));
}
}
}
// 若开启Dubbo服务检查,并且服务接口不可用,则直接抛异常
if (shouldCheck() && !invoker.isAvailable()) {
throw new IllegalStateException("Failed to check the status of the service " + interfaceName + ". No provider available for the service " + (group == null ? "" : group + "/") + interfaceName + (version == null ? "" : ":" + version) + " from the url " + invoker.getUrl() + " to the consumer " + NetUtils.getLocalHost() + " use dubbo version " + Version.getVersion());
}
if (logger.isInfoEnabled()) {
logger.info("Refer dubbo service " + interfaceClass.getName() + " from url " + invoker.getUrl());
}
/**
* @since 2.7.0
* ServiceData Store
*/
MetadataReportService metadataReportService = null;
if ((metadataReportService = getMetadataReportService()) != null) {
URL consumerURL = new URL(CONSUMER_PROTOCOL, map.remove(REGISTER_IP_KEY), 0, map.get(INTERFACE_KEY), map);
metadataReportService.publishConsumer(consumerURL);// 9.消费者元数据信息发布
}
// create service proxy
return (T) PROXY_FACTORY.getProxy(invoker);// 10.真正开始创建代理对象,通过代理工厂创建代理引用(代理工厂的默认扩展点名是 javassist )
}
上述方法中有几处重要得地方:1、2、3、4、5、6(单节点)、7、8(集群)、9、10;其中6(7)、8、10是重中之重。下面主要来分析6(7)、8、10这几个关键步骤:
6(7) 步骤核心子流程如下:
① 调用协议的自适应扩展点获取真正端得协议扩展点(Registryprotocol),(扩展点的原理:Dubbo设计之ExtensionLoader),但是由于Protocol扩展点有包装类(ProtocolFilterWrapper、ProtocolListenerWrapper)
,所以获取真正的扩展实现之后会被包装类包装起来: ProtocolFilterWrapper.refer -> ProtocolListenerWrapper.refer -> Registryprotocol.refer :
// ProtocolFilterWrapper
@Override
public Invoker refer(Class type, URL url) throws RpcException {
if (REGISTRY_PROTOCOL.equals(url.getProtocol())) {
return protocol.refer(type, url); // 由于当前协议是registry,调用ProtocolListenerWrapper的 refer方法
}
return buildInvokerChain(protocol.refer(type, url), REFERENCE_FILTER_KEY, CommonConstants.CONSUMER);
}
// ProtocolListenerWrapper
@Override
public Invoker refer(Class type, URL url) throws RpcException {
if (REGISTRY_PROTOCOL.equals(url.getProtocol())) {
return protocol.refer(type, url); // 由于当前协议是registry,调用 RegistryProtocol 的refer方法
}
return new ListenerInvokerWrapper(protocol.refer(type, url),
Collections.unmodifiableList(
ExtensionLoader.getExtensionLoader(InvokerListener.class)
.getActivateExtension(url, INVOKER_LISTENER_KEY)));
}
② 接着通过ProtocolListenerWrapper调用真实的扩展实现RegistryProtocol 的refer方法:
// RegistryProtocol
@Override
@SuppressWarnings("unchecked")
public Invoker refer(Class type, URL url) throws RpcException {
// 重新构建一个protocol为 `zookeeper`的url,如果url没有指定protocol的话,默认是 `dubbo`
url = URLBuilder.from(url)
.setProtocol(url.getParameter(REGISTRY_KEY, DEFAULT_REGISTRY))
.removeParameter(REGISTRY_KEY)
.build();
// 调用 `RegistryFactory扩展点`的自适应扩展实现,此处根据url中的protocol=zookeeper,会去找到真正的扩展点实现`ZookeeperRegistryFactory`,然后调用其getRegistry方法(实际上此方法继承自父类 AbstractRegistryFactory)生成ZookeeperRegistry实例。
Registry registry = registryFactory.getRegistry(url);
if (RegistryService.class.equals(type)) {
return proxyFactory.getInvoker((T) registry, type, url);
}
// group="a,b" or group="*"
Map qs = StringUtils.parseQueryString(url.getParameterAndDecoded(REFER_KEY));
String group = qs.get(GROUP_KEY);// 假设group为空的情况
if (group != null && group.length() > 0) {
if ((COMMA_SPLIT_PATTERN.split(group)).length > 1 || "*".equals(group)) {
return doRefer(getMergeableCluster(), registry, type, url);// 不执行此处代码
}
}
return doRefer(cluster, registry, type, url);// 通过集群扩展点对注册实例进行包装
}
③ 调用 ZookeeperRegistryFactory 的 getRegistry方法(继承自 AbstractRegistryFactory )获取注册实现:
// AbstractRegistryFactory
@Override
public Registry getRegistry(URL url) {
// 构建一个路径为RegistryService的注册URL
url = URLBuilder.from(url)
.setPath(RegistryService.class.getName())
.addParameter(INTERFACE_KEY, RegistryService.class.getName())
.removeParameters(EXPORT_KEY, REFER_KEY)
.build();
String key = url.toServiceStringWithoutResolving();
// Lock the registry access process to ensure a single instance of the registry
LOCK.lock();
try {
Registry registry = REGISTRIES.get(key);// 从缓存中取,不等于空直接返回
if (registry != null) {
return registry;
}
//create registry by spi/ioc
registry = createRegistry(url); // 模板方法,通过具体的实现类(此处是ZookeeperRegistryFactory)去获取注册实例
if (registry == null) {
throw new IllegalStateException("Can not create registry " + url);
}
REGISTRIES.put(key, registry); // 获取之后放入缓存中,防止并发场景下重复创建
return registry;
} finally {
// Release the lock
LOCK.unlock();
}
}
protected abstract Registry createRegistry(URL url); // 抽象模板方法
// ZookeeperRegistryFactory 的方法实现
@Override
public Registry createRegistry(URL url) {
// 创建注册实例,注册的url为 zookeeper://xxx:2181/org.apache.dubbo.registry.RegistryService?application=dubbo-api-consumer&dubbo=XXX&interface=org.apache.dubbo.registry.RegistryService&pid=13740×tamp=1640015273548
return new ZookeeperRegistry(url, zookeeperTransporter);
}
// 构造器
public ZookeeperRegistry(URL url, ZookeeperTransporter zookeeperTransporter) {
super(url);
if (url.isAnyHost()) {
throw new IllegalStateException("registry address == null");
}
String group = url.getParameter(GROUP_KEY, DEFAULT_ROOT);
if (!group.startsWith(PATH_SEPARATOR)) {
group = PATH_SEPARATOR + group;
}
this.root = group;
zkClient = zookeeperTransporter.connect(url);// 连接zk server端
zkClient.addStateListener(state -> {
if (state == StateListener.RECONNECTED) {
try {
recover(); // 故障恢复
} catch (Exception e) {
logger.error(e.getMessage(), e);
}
}
});
}
// FailbackRegistry( ZookeeperRegistry的父类 )
@Override
protected void recover() throws Exception {
// register
Set recoverRegistered = new HashSet(getRegistered());// 获取已经注册的URL
if (!recoverRegistered.isEmpty()) {
if (logger.isInfoEnabled()) {
logger.info("Recover register url " + recoverRegistered);
}
for (URL url : recoverRegistered) {
addFailedRegistered(url);// 添加到失败的列表中,并通过时间轮(HashedWheelTimer)定时重试
}
}
// subscribe
Map> recoverSubscribed = new HashMap>(getSubscribed()); // 获取订阅的URL跟对应的通知状态变更的监听器
if (!recoverSubscribed.isEmpty()) {
if (logger.isInfoEnabled()) {
logger.info("Recover subscribe url " + recoverSubscribed.keySet());
}
for (Map.Entry> entry : recoverSubscribed.entrySet()) {
URL url = entry.getKey();
for (NotifyListener listener : entry.getValue()) {
addFailedSubscribed(url, listener);// 添加到失败订阅列表中,并通过时间轮(HashedWheelTimer)定时重试
}
}
}
}
④ 调用 org.apache.dubbo.registry.integration.RegistryProtocol#doRefer方法,通过集群扩展点Cluster
对本地注册目录进行合并:
private Invoker doRefer(Cluster cluster, Registry registry, Class type, URL url) {
// 创建本地注册目录(RegistryDirectory本身就是一个org.apache.dubbo.registry.NotifyListener),当监听到服务端地址变化时,更新本地url信息
RegistryDirectory directory = new RegistryDirectory(type, url);
directory.setRegistry(registry);
directory.setProtocol(protocol);
// all attributes of REFER_KEY
Map parameters = new HashMap(directory.getUrl().getParameters());
URL subscribeUrl = new URL(CONSUMER_PROTOCOL, parameters.remove(REGISTER_IP_KEY), 0, type.getName(), parameters);
if (!ANY_VALUE.equals(url.getServiceInterface()) && url.getParameter(REGISTER_KEY, true)) {
// 设置消费的URL
directory.setRegisteredConsumerUrl(getRegisteredConsumerUrl(subscribeUrl, url));
// 调用注册实例的注册方法,将消费的URL注册到本地
registry.register(directory.getRegisteredConsumerUrl());
}
// 构建路由链 (根据优先级依次为 MockRouterFactory、TagRouterFactor、AppRouterFactory、ServiceRouterFactory)
directory.buildRouterChain(subscribeUrl);
// 订阅URL(consumer://127.0..1/com.alibaba.mp.alipay.AlipayInfoAbility?application=dubbo-demo-api-consumer&category=providers,configurators,routers&dubbo=2.0.2&interface=com.alibaba.mp.alipay.AlipayInfoAbility&lazy=false&methods=getAlipayInfo&pid=11856&side=consumer&sticky=false×tamp=1640020348831)
directory.subscribe(subscribeUrl.addParameter(CATEGORY_KEY,
PROVIDERS_CATEGORY + "," + CONFIGURATORS_CATEGORY + "," + ROUTERS_CATEGORY));
// 合并成一个virtual的Invoker
Invoker invoker = cluster.join(directory);
ProviderConsumerRegTable.registerConsumer(invoker, url, subscribeUrl, directory);
return invoker;
}
⑤ 注册消费的URL、构建调用路由链、订阅要消费的URL(RegistryDirectory中)
// FailbackRegistry 重写了AbstractRegistry的register方法,增加将消费者URL注册到服务端的的逻辑
@Override
public void register(URL url) {
// 缓存到本地
super.register(url);
removeFailedRegistered(url);
removeFailedUnregistered(url);
try {
// Sending a registration request to the server side
doRegister(url); // 模板方法,向服务端发送注册请求
} catch (Exception e) {
Throwable t = e;
...
}
}
// ZookeeperRegistry
@Override
public void doRegister(URL url) {
try {
// 在zk上创建提供方URL的临时节点(提供方目录是临时的)
zkClient.create(toUrlPath(url), url.getParameter(DYNAMIC_KEY, true));
} catch (Throwable e) {
...
}
}
// 构建路由链,用于后续从集群中选择路由时使用
public void buildRouterChain(URL url) {
this.setRouterChain(RouterChain.buildChain(url));
}
// RouterChain
public static RouterChain buildChain(URL url) {
return new RouterChain<>(url);
}
private RouterChain(URL url) {
// 自然序获取RouterFactory所有的激活扩展点实现,(MockRouterFactory、TagRouterFactor、AppRouterFactory、ServiceRouterFactory)
List extensionFactories = ExtensionLoader.getExtensionLoader(RouterFactory.class)
.getActivateExtension(url, (String[]) null);
// 获取路由集合
List routers = extensionFactories.stream()
.map(factory -> factory.getRouter(url))
.collect(Collectors.toList());
initWithRouters(routers);
}
// RegistryDirectory 的订阅方法
public void subscribe(URL url) {
// 缓存到本地注册目录中
setConsumerUrl(url);
// 添加 节点变更监听器用于通知服务上下线
CONSUMER_CONFIGURATION_LISTENER.addNotifyListener(this);
// 通知 configurator/router 变更
serviceConfigurationListener = new ReferenceConfigurationListener(this, url);
// 开始订阅消费URL,每个对应一个注册目录,并注册与远端服务子节点之间的映射关系,当zk节点发生变化时会回调当前监听器更新本地消费URL
registry.subscribe(url, this);
}
// FailbackRegistry,listener为 RegistryDirectory
@Override
public void subscribe(URL url, NotifyListener listener) {
super.subscribe(url, listener);
removeFailedSubscribed(url, listener);
try {
// Sending a subscription request to the server side
doSubscribe(url, listener); // 模板方法,调用子类(ZookeeperRegistry)的订阅方法
} catch (Exception e) {
Throwable t = e;
...
}
}
// ZookeeperRegistry 的模板实现方法
@Override
public void doSubscribe(final URL url, final NotifyListener listener) {
try {
if (ANY_VALUE.equals(url.getServiceInterface())) {
...
} else {
List urls = new ArrayList<>();
// toCategoriesPath:
// 0 = "/dubbo/com.alibaba.mp.alipay.AlipayInfoAbility/providers"
// 1 = "/dubbo/com.alibaba.mp.alipay.AlipayInfoAbility/configurators"
// 2 = "/dubbo/com.alibaba.mp.alipay.AlipayInfoAbility/routers"
for (String path : toCategoriesPath(url)) {
ConcurrentMap listeners = zkListeners.get(url);
if (listeners == null) {
zkListeners.putIfAbsent(url, new ConcurrentHashMap<>());
listeners = zkListeners.get(url);
}
ChildListener zkListener = listeners.get(listener);
if (zkListener == null) {
zkListener = listeners.get(listener);
// 创建zk子节点监听器,用于监听节点变化,然后调用notify()方法变更本地目录
listeners.putIfAbsent(listener, (parentPath, currentChilds) -> ZookeeperRegistry.this.notify(url, listener, toUrlsWithEmpty(url, parentPath, currentChilds)));
}
// 创建持久化节点
zkClient.create(path, false);
List children = zkClient.addChildListener(path, zkListener);
if (children != null) {
urls.addAll(toUrlsWithEmpty(url, path, children));
}
}
// 通知RegistryDirectory监听器,修改本地URL
notify(url, listener, urls);
}
} catch (Throwable e) {
...
}
}
// FailbackRegistry
@Override
protected void notify(URL url, NotifyListener listener, List urls) {
if (url == null) {
throw new IllegalArgumentException("notify url == null");
}
if (listener == null) {
throw new IllegalArgumentException("notify listener == null");
}
try {
doNotify(url, listener, urls); // 模板方法,ZookeeperRegistry 实现通知变更
} catch (Exception t) {
...
}
}
// AbstractRegistry
protected void notify(URL url, NotifyListener listener, List urls) {
if (url == null) {
throw new IllegalArgumentException("notify url == null");
}
if (listener == null) {
throw new IllegalArgumentException("notify listener == null");
}
if ((CollectionUtils.isEmpty(urls))
&& !ANY_VALUE.equals(url.getServiceInterface())) {
logger.warn("Ignore empty notify urls for subscribe url " + url);
return;
}
if (logger.isInfoEnabled()) {
logger.info("Notify urls for subscribe url " + url + ", urls: " + urls);
}
// keep every provider's category.
Map> result = new HashMap<>();
for (URL u : urls) {
if (UrlUtils.isMatch(url, u)) {
String category = u.getParameter(CATEGORY_KEY, DEFAULT_CATEGORY);
List categoryList = result.computeIfAbsent(category, k -> new ArrayList<>());
categoryList.add(u);
}
}
if (result.size() == 0) {
return;
}
// 将提供方的URL放入到已通知的对象映射中
Map> categoryNotified = notified.computeIfAbsent(url, u -> new ConcurrentHashMap<>());
for (Map.Entry> entry : result.entrySet()) {
String category = entry.getKey();
List categoryList = entry.getValue();
categoryNotified.put(category, categoryList);
// 调用监听器通知变更,listener = RegistryDirectory
listener.notify(categoryList);
// We will update our cache file after each notification.
// When our Registry has a subscribe failure due to network jitter, we can return at least the existing cache URL.
saveProperties(url);
}
}
// RegistryDirectory,更新本地对应的目录
@Override
public synchronized void notify(List urls) {
// 根据 configurators、routers、providers进行分组
Map> categoryUrls = urls.stream()
.filter(Objects::nonNull)
.filter(this::isValidCategory)
.filter(this::isNotCompatibleFor26x)
.collect(Collectors.groupingBy(url -> {
if (UrlUtils.isConfigurator(url)) {
return CONFIGURATORS_CATEGORY;
} else if (UrlUtils.isRoute(url)) {
return ROUTERS_CATEGORY;
} else if (UrlUtils.isProvider(url)) {
return PROVIDERS_CATEGORY;
}
return "";
}));
// 若监听到category有变更,则取代之前的值
List configuratorURLs = categoryUrls.getOrDefault(CONFIGURATORS_CATEGORY, Collections.emptyList());
this.configurators = Configurator.toConfigurators(configuratorURLs).orElse(this.configurators);
// 若监听到有新得路由,则添加到路由链中
List routerURLs = categoryUrls.getOrDefault(ROUTERS_CATEGORY, Collections.emptyList());
toRouters(routerURLs).ifPresent(this::addRouters);
// providers
List providerURLs = categoryUrls.getOrDefault(PROVIDERS_CATEGORY, Collections.emptyList());
// 刷新+覆盖
refreshOverrideAndInvoker(providerURLs);
}
private void refreshOverrideAndInvoker(List urls) {
// mock zookeeper://xxx?mock=return null
overrideDirectoryUrl();
// 刷新本地缓存的Invokers
refreshInvoker(urls);
}
private void refreshInvoker(List invokerUrls) {
Assert.notNull(invokerUrls, "invokerUrls should not be null");
if (invokerUrls.size() == 1
&& invokerUrls.get(0) != null
&& EMPTY_PROTOCOL.equals(invokerUrls.get(0).getProtocol())) {
this.forbidden = true; // Forbid to access
this.invokers = Collections.emptyList();
routerChain.setInvokers(this.invokers);
destroyAllInvokers(); // Close all invokers
} else {
this.forbidden = false; // Allow to access
Map> oldUrlInvokerMap = this.urlInvokerMap; // local reference
if (invokerUrls == Collections.emptyList()) {
invokerUrls = new ArrayList<>();
}
if (invokerUrls.isEmpty() && this.cachedInvokerUrls != null) {
invokerUrls.addAll(this.cachedInvokerUrls);
} else {
this.cachedInvokerUrls = new HashSet<>();
this.cachedInvokerUrls.addAll(invokerUrls);//Cached invoker urls, convenient for comparison
}
if (invokerUrls.isEmpty()) {
return;
}
Map> newUrlInvokerMap = toInvokers(invokerUrls);// Translate url list to Invoker map
/**
* If the calculation is wrong, it is not processed.
*
* 1. The protocol configured by the client is inconsistent with the protocol of the server.
* eg: consumer protocol = dubbo, provider only has other protocol services(rest).
* 2. The registration center is not robust and pushes illegal specification data.
*
*/
if (CollectionUtils.isEmptyMap(newUrlInvokerMap)) {
logger.error(new IllegalStateException("urls to invokers error .invokerUrls.size :" + invokerUrls.size() + ", invoker.size :0. urls :" + invokerUrls
.toString()));
return;
}
List> newInvokers = Collections.unmodifiableList(new ArrayList<>(newUrlInvokerMap.values()));
// pre-route and build cache, notice that route cache should build on original Invoker list.
// toMergeMethodInvokerMap() will wrap some invokers having different groups, those wrapped invokers not should be routed.
routerChain.setInvokers(newInvokers);
// 覆盖本地缓存,调用的时候就是从此成员变量中选取的
this.invokers = multiGroup ? toMergeInvokerList(newInvokers) : newInvokers;
this.urlInvokerMap = newUrlInvokerMap;
try {
destroyUnusedInvokers(oldUrlInvokerMap, newUrlInvokerMap); // Close the unused Invoker
} catch (Exception e) {
logger.warn("destroyUnusedInvokers error. ", e);
}
}
}
private Map> toInvokers(List urls) {
Map> newUrlInvokerMap = new HashMap<>();
if (urls == null || urls.isEmpty()) {
return newUrlInvokerMap;
}
Set keys = new HashSet<>();
String queryProtocols = this.queryMap.get(PROTOCOL_KEY);
for (URL providerUrl : urls) {
// If protocol is configured at the reference side, only the matching protocol is selected
if (queryProtocols != null && queryProtocols.length() > 0) {
boolean accept = false;
String[] acceptProtocols = queryProtocols.split(",");
for (String acceptProtocol : acceptProtocols) {
if (providerUrl.getProtocol().equals(acceptProtocol)) {
accept = true;
break;
}
}
if (!accept) {
continue;
}
}
if (EMPTY_PROTOCOL.equals(providerUrl.getProtocol())) {
continue;
}
// protocol = dubbo
if (!ExtensionLoader.getExtensionLoader(Protocol.class).hasExtension(providerUrl.getProtocol())) {
logger.error(new IllegalStateException("Unsupported protocol " + providerUrl.getProtocol() +
" in notified url: " + providerUrl + " from registry " + getUrl().getAddress() +
" to consumer " + NetUtils.getLocalHost() + ", supported protocol: " +
ExtensionLoader.getExtensionLoader(Protocol.class).getSupportedExtensions()));
continue;
}
URL url = mergeUrl(providerUrl);
String key = url.toFullString(); // The parameter urls are sorted
if (keys.contains(key)) { // Repeated url
continue;
}
keys.add(key);
// Cache key is url that does not merge with consumer side parameters, regardless of how the consumer combines parameters, if the server url changes, then refer again
Map> localUrlInvokerMap = this.urlInvokerMap; // local reference
Invoker invoker = localUrlInvokerMap == null ? null : localUrlInvokerMap.get(key);
if (invoker == null) { // Not in the cache, refer again
try {
boolean enabled = true;
if (url.hasParameter(DISABLED_KEY)) {
enabled = !url.getParameter(DISABLED_KEY, false);
} else {
enabled = url.getParameter(ENABLED_KEY, true);
}
// 默认情况下,enabled = true
if (enabled) {
// 根据当前key创建一个委派Invoker:InvokerDelegate。
// 重点:而此处url的protocol = dubbo,那么 protocol.refer(serviceType, url)会执行 Dubboprotocol的refer方法
// 由于此处的 protocol引用是自适应扩展点,所以还是会经过ProtocolFilterWrapper跟ProtocolListenerWrapper进行包装:ProtocolFilterWrapper.refer -> ProtocolListenerWrapper.refer -> Dubboprotocol.refer
invoker = new InvokerDelegate<>(protocol.refer(serviceType, url), url, providerUrl);
}
} catch (Throwable t) {
logger.error("Failed to refer invoker for interface:" + serviceType + ",url:(" + url + ")" + t.getMessage(), t);
}
if (invoker != null) { // Put new invoker in cache
newUrlInvokerMap.put(key, invoker);
}
} else {
newUrlInvokerMap.put(key, invoker);
}
}
keys.clear();
return newUrlInvokerMap;
}
// ProtocolFilterWrapper 的refer方法
@Override
public Invoker refer(Class type, URL url) throws RpcException {
if (REGISTRY_PROTOCOL.equals(url.getProtocol())) {// 协议是dubbo,此处不成立
return protocol.refer(type, url);
}
// 调用 ProtocolListenerWrapper 的refer方法先生成一个Invoker
// 将生成的Invoker与Filter扩展点构建消费端的Invoker链,REFERENCE_FILTER_KEY = reference.filter,
// 经过返回 ProtocolFilterWrapper 之后,生成的是 CallbackRegistrationInvoker
return buildInvokerChain(protocol.refer(type, url), REFERENCE_FILTER_KEY, CommonConstants.CONSUMER);
}
// 构建Filter链
private static Invoker buildInvokerChain(final Invoker invoker, String key, String group) {
// 此处的 invoker是ProtocolListenerWrapper.refer执行之后的Invoker,即 ListenerInvokerWrapper
Invoker last = invoker;
// 获取消费者端的所有满足条件的激活扩展点,此处会根据排序字段升序排列
// ConsumerContextFilter -> FutureFilter -> MonitorFilter
List filters = ExtensionLoader.getExtensionLoader(Filter.class).getActivateExtension(invoker.getUrl(), key, group);
if (!filters.isEmpty()) {
// 循环从后往前组成Invoker链,为了调用时按照设定的顺序执行
// 循环顺序:MonitorFilter -> FutureFilter -> ConsumerContextFilter
// 请求处理顺序:ConsumerContextFilter -> FutureFilter -> MonitorFilter
for (int i = filters.size() - 1; i >= 0; i--) {
final Filter filter = filters.get(i);
final Invoker next = last;
last = new Invoker() {
...
...
@Override
public Result invoke(Invocation invocation) throws RpcException {
Result asyncResult;
try {
asyncResult = filter.invoke(next, invocation);
} catch (Exception e) {
// onError callback
if (filter instanceof ListenableFilter) {
Filter.Listener listener = ((ListenableFilter) filter).listener();
if (listener != null) {
// 钩子方法
listener.onError(e, invoker, invocation);
}
}
throw e;
}
return asyncResult;
}
@Override
public void destroy() {
invoker.destroy();
}
@Override
public String toString() {
return invoker.toString();
}
};
}
}
// 创建一个回调的 Invoker,如果定义的Filter是 ListenableFilter的实现类
// 那么调用返回时,会回调Filter.Listener的 onResponse(正常返回) 和 onError方法(异常情况)
return new CallbackRegistrationInvoker<>(last, filters);
}
// ProtocolListenerWrapper的refer方法
@Override
public Invoker refer(Class type, URL url) throws RpcException {
if (REGISTRY_PROTOCOL.equals(url.getProtocol())) {// protocol = dubbo,此处不成立
return protocol.refer(type, url);
}
// 此处的 protocol是真实的 DubboProtocol
// 调用 DubboProtocol的refer 之后生成 DubboInvoker
// 获取 InvokerListener的激活扩展点实现,若是存在,则调用扩展实现的 referred(Invoker> invoker)方法,(类似Spring中的BeanPostProcessor)
// 返回 ListenerInvokerWrapper
return new ListenerInvokerWrapper(protocol.refer(type, url),
Collections.unmodifiableList(
ExtensionLoader.getExtensionLoader(InvokerListener.class)
.getActivateExtension(url, INVOKER_LISTENER_KEY)));
}
// AbstractProtocol (DubboProtocol的父类)
@Override
public Invoker refer(Class type, URL url) throws RpcException {
// protocolBindingRefer 方法是个抽象方法,让其子类去实现
return new AsyncToSyncInvoker<>(protocolBindingRefer(type, url));
}
// DubboProtocol
@Override
public Invoker protocolBindingRefer(Class serviceType, URL url) throws RpcException {
optimizeSerialization(url);
// create rpc invoker.
DubboInvoker invoker = new DubboInvoker(serviceType, url, getClients(url), invokers);
invokers.add(invoker);
return invoker;
}
private ExchangeClient[] getClients(URL url) {
// whether to share connection
boolean useShareConnect = false;
int connections = url.getParameter(CONNECTIONS_KEY, 0);
List shareClients = null;
// if not configured, connection is shared, otherwise, one connection for one service
// 没有配置 connections 参数的话,默认使用共享连接(一个服务公用一个连接)
if (connections == 0) {
useShareConnect = true;
/**
* The xml configuration should have a higher priority than properties.
*/
String shareConnectionsStr = url.getParameter(SHARE_CONNECTIONS_KEY, (String) null);
connections = Integer.parseInt(StringUtils.isBlank(shareConnectionsStr) ? ConfigUtils.getProperty(SHARE_CONNECTIONS_KEY,
DEFAULT_SHARE_CONNECTIONS) : shareConnectionsStr);
// 获取共享客户端连接
shareClients = getSharedClient(url, connections);
}
// 根据连接数创建请求客户端数组(ReferenceCountExchangeClient)
ExchangeClient[] clients = new ExchangeClient[connections];
for (int i = 0; i < clients.length; i++) {
// 使用共享连接
if (useShareConnect) {
clients[i] = shareClients.get(i); // shareconnections 优先级高于 connections
} else {
clients[i] = initClient(url); // 初始化客户端连接 HeaderExchangeClient
}
}
return clients;
}
private ReferenceCountExchangeClient buildReferenceCountExchangeClient(URL url) {
// 检查完缓存中没有的话,也会调用此方法去创建客户端连接
ExchangeClient exchangeClient = initClient(url);
return new ReferenceCountExchangeClient(exchangeClient);
}
private ExchangeClient initClient(URL url) {
// client type setting.
String str = url.getParameter(CLIENT_KEY, url.getParameter(SERVER_KEY, DEFAULT_REMOTING_CLIENT));
// 设置默认的编解码为 codec = dubbo(DubboCodec)
url = url.addParameter(CODEC_KEY, DubboCodec.NAME);
// enable heartbeat by default(1分钟)
url = url.addParameterIfAbsent(HEARTBEAT_KEY, String.valueOf(DEFAULT_HEARTBEAT));
// BIO is not allowed since it has severe performance issue.
if (str != null && str.length() > 0 && !ExtensionLoader.getExtensionLoader(Transporter.class).hasExtension(str)) {
throw new RpcException("Unsupported client type: " + str + "," +
" supported client type is " + StringUtils.join(ExtensionLoader.getExtensionLoader(Transporter.class).getSupportedExtensions(), " "));
}
ExchangeClient client;
try {
// connection should be lazy
if (url.getParameter(LAZY_CONNECT_KEY, false)) {
client = new LazyConnectExchangeClient(url, requestHandler);
} else {
// 通过交换层 Exchangers工具类 生成客户端
client = Exchangers.connect(url, requestHandler);
}
} catch (RemotingException e) {
throw new RpcException("Fail to create remoting client for service(" + url + "): " + e.getMessage(), e);
}
return client;
}
// 中间的代码就不一一粘贴了,下面来看一下创建NettyClient时关键性的一步:
/**
* The constructor of NettyClient.
* It wil init and start netty.
*/
public NettyClient(final URL url, final ChannelHandler handler) throws RemotingException {
// 调用父类构造器
super(url, wrapChannelHandler(url, handler));
}
// AbstractClient 的构造器,有几处模板方法:doOpen()、connect()、close()
public AbstractClient(URL url, ChannelHandler handler) throws RemotingException {
super(url, handler);
needReconnect = url.getParameter(Constants.SEND_RECONNECT_KEY, false);
try {
// 打开连接通道
doOpen();
} catch (Throwable t) {
close();
...
}
try {
// connect.
connect();
...
}
} catch (RemotingException t) {
if (url.getParameter(Constants.CHECK_KEY, true)) {
close(); // 异常关闭连接
throw t;
} else {
...
}
} catch (Throwable t) {
close(); // 异常关闭连接
...
}
...
}
// NettyClient(注意此处说的这个类都是Dubbo自己的,Netty里面可没有叫这玩意的)抽象实现
@Override
protected void doOpen() throws Throwable {
// 创建一个统一出入站处理器(为啥是出入? 因为实现了 io.netty.channel.ChannelDuplexHandler)
final NettyClientHandler nettyClientHandler = new NettyClientHandler(getUrl(), this);
// 创建客户端启动类
bootstrap = new Bootstrap();
// nioEventLoopGroup 是一个线程数为 CPU+1的 NioEventLoopGroup(事件循环组,可以当成线程池)
bootstrap.group(nioEventLoopGroup)
.option(ChannelOption.SO_KEEPALIVE, true)
.option(ChannelOption.TCP_NODELAY, true)
.option(ChannelOption.ALLOCATOR, PooledByteBufAllocator.DEFAULT)
//.option(ChannelOption.CONNECT_TIMEOUT_MILLIS, getTimeout())
// io.netty.channel.socket.nio.NioSocketChannel中会缓存一个 SocketChannel 用于与服务端进行通信
.channel(NioSocketChannel.class);
bootstrap.option(ChannelOption.CONNECT_TIMEOUT_MILLIS, Math.max(3000, getConnectTimeout()));
bootstrap.handler(new ChannelInitializer() {
@Override
protected void initChannel(Channel ch) throws Exception {
int heartbeatInterval = UrlUtils.getHeartbeat(getUrl());
// 自定义编解码器(粘包拆包)
NettyCodecAdapter adapter = new NettyCodecAdapter(getCodec(), getUrl(), NettyClient.this);
ch.pipeline()
// 可以参考 io.netty.channel.ChannelPipeline 中的注释说明
// 读:decoder -> client-idle-handler -> handler
// 写:handler -> client-idle-handler -> encoder
.addLast("decoder", adapter.getDecoder())
.addLast("encoder", adapter.getEncoder())
.addLast("client-idle-handler", new IdleStateHandler(heartbeatInterval, 0, 0, MILLISECONDS))
.addLast("handler", nettyClientHandler);
String socksProxyHost = ConfigUtils.getProperty(SOCKS_PROXY_HOST);
if(socksProxyHost != null) {
int socksProxyPort = Integer.parseInt(ConfigUtils.getProperty(SOCKS_PROXY_PORT, DEFAULT_SOCKS_PROXY_PORT));
Socks5ProxyHandler socks5ProxyHandler = new Socks5ProxyHandler(new InetSocketAddress(socksProxyHost, socksProxyPort));
ch.pipeline().addFirst(socks5ProxyHandler);
}
}
});
}
// NettyClient 抽象实现
@Override
protected void doConnect() throws Throwable {
long start = System.currentTimeMillis();
// 通过调用netty 封装好的客户端与服务端进行连接,此连接会立马返回一个 ChannelFuture(DefaultChannelPromise)。
// 类似于 JUC 中的 Future
ChannelFuture future = bootstrap.connect(getConnectAddress());
try {
// 根据超时时间不中断获取连接结果
boolean ret = future.awaitUninterruptibly(getConnectTimeout(), MILLISECONDS);
// 连接成功
if (ret && future.isSuccess()) {
Channel newChannel = future.channel();
try {
// Close old channel
// copy reference
Channel oldChannel = NettyClient.this.channel;
if (oldChannel != null) {
try {
if (logger.isInfoEnabled()) {
logger.info("Close old netty channel " + oldChannel + " on create new netty channel " + newChannel);
}
oldChannel.close();
} finally {
NettyChannel.removeChannelIfDisconnected(oldChannel);
}
}
} finally {
if (NettyClient.this.isClosed()) {
try {
if (logger.isInfoEnabled()) {
logger.info("Close new netty channel " + newChannel + ", because the client closed.");
}
newChannel.close();
} finally {
NettyClient.this.channel = null;
NettyChannel.removeChannelIfDisconnected(newChannel);
}
} else {
NettyClient.this.channel = newChannel;
}
}
} else if (future.cause() != null) {
...
} finally {
// just add new valid channel to NettyChannel's cache
if (!isConnected()) {
//future.cancel(true);
}
}
}
关键流程:
- Invoker链创建过程:
① DubboInvoker -> 真正发送请求的Invoker
② AsyncToSyncInvoker(DubboInvoker ** ) ->
③ ListenerInvokerWrapper( AsyncToSyncInvoker ) ->
④ CallbackRegistrationInvoker( ListenerInvokerWrapper ) -> Filter扩展点钩子回调
⑤ InvokerDelegate(CallbackRegistrationInvoker) ->
⑥ FailoverClusterInvoker( RegistryDirectory(InvokerDelegates) ) -> RegistryDirectory 中存放调用时的Invokes
⑦ MockClusterInvoker (RegistryDirectory,FailoverClusterInvoker**) 暴漏给接口拦截的入口Invoker
- PS:上面是创建顺序,调用时自然是相反的顺序。那么如何方便记忆呢 ?
Ⅰ 最后一个Invoker,实际发出网络请求的Invoker:DubboInvoker,铁打不变 !!!
Ⅱ 由于 DubboInvoker 是异步返回,若是客户端指定同步调用(默认同步),则需要在AsyncToSyncInvoker(最靠近DubboInvoker)
Ⅲ CallbackRegistrationInvoker,对返回结果的第一层处理。通过定义的Filter类型决定是否调用 onResponse方法 或者 onError方法
Ⅳ Mock优先
Ⅴ 高可用高于任何其他的执行Invoker( 故障转移:FailoverClusterInvoker )
- Netty客户端创建过程:
org.apache.dubbo.rpc.protocol.dubbo.DubboProtocol#requestHandler -> 真正处理消息的收发处理器
HeaderExchangeHandler(requestHandler) ->
DecodeHandler(HeaderExchangeHandler) -> 自定义编解码处理器(拆包、粘包)
AllChannelHandler(DecodeHandler,URL) -> 请求分发处理器,默认所有请求都转发
HeartbeatHandler(AllChannelHandler) ->
MultiMessageHandler (HeartbeatHandler) -> 消息类型是MultiMessage
,不过一般的请求都是Request
类型
NettyClient(Url,MultiMessageHandler) -> 封装了一系列打开、链接通道逻辑
HeaderExchangeClient(NettyClient,true) ->
ReferenceCountExchangeClient(HeaderExchangeClient) 共享连接才会创建 ReferenceCountExchangeClient,暴漏给客户端使用的Client
8这一步是通过如下两行代码:
// 设置cluster = registryaware( RegistryAwareCluster )
URL u = registryURL.addParameter(CLUSTER_KEY, RegistryAwareCluster.NAME);
// 通过集群的自适应扩展点将多个Invokers( MockClusterInvoker 集合)合并成一个Invoker(RegistryAwareClusterInvoker是用来做集群路由,选择一个实际调用的Invoker),此处RegistryAwareClusterInvoker是用来做集群路由,选择一个实际调用的Invoker
invoker = CLUSTER.join(new StaticDirectory(u, invokers))
到这里,大家千万不要被 "满眼的Invoker" 给绕晕了,仔细分析一下Dubbo的执行链与模块分层是可以理清楚的。
接着上面 执行链,此处会先包装一个StaticDirectory,然后 :
MockClusterWrapper(MockClusterInvoker )-> RegistryAwareCluster(RegistryAwareClusterInvoker, 最终返回的Invoker)-> StaticDirectory (URL , List)
StaticDirectory:通过前面设置的Router链来执行路由选择,会选择最后一个Router所路由到的List集合上
LoadBalance:负载均衡器(默认random,随机),从前面选择到的List集合上,根据负载均衡算法选择一个
( 路由跟负载均衡都是调用invoke方法时才会用到,我们初始化过程分析完之后,调用过程自然就清晰了 )
9步骤有以下核心子流程:
① 接着就开始创建代理引用了: 通过调用 ProxyFactory
扩展点的适配类的getProxy方法获取代理对象:
// ProxyFactory$Adaptive.getProxy(invoker) -> 找到默认的扩展点名称(javassist)-> JavassistProxyFactory(实际是父类 AbstractProxyFactory的 getProxy方法)
return (T) PROXY_FACTORY.getProxy(invoker)
// AbstractProxyFactory
@Override
public T getProxy(Invoker invoker) throws RpcException {
// 非泛化方式获取接口代理对象
return getProxy(invoker, false);
}
// JavassistProxyFactory
@Override
@SuppressWarnings("unchecked")
public T getProxy(Invoker invoker, Class>[] interfaces) {
// 此处的创建代理的方式类似于 JDK动态代理,不过生成的class是借助与 javassist 字节码工具自定义的
// 而且生成的代理类继承了org.apache.dubbo.common.bytecode.Proxy
return (T) Proxy.getProxy(interfaces).newInstance(new InvokerInvocationHandler(invoker));
}
② 若不指定 interfaces这个key的话,生成的代理类会同时实现interface
跟 EchoService(用于回声测试)
接口,具体代码 Proxy.getProxy(interfaces).newInstance(new InvokerInvocationHandler(invoker))
。其中 Proxy.getProxy(interfaces) 会为接口生成一个代理类,代理继承了 org.apache.dubbo.common.bytecode.Proxy
抽象类并且实现了 interfaces
。生成的代理类如下:
public class com.alibaba.dubbo.common.bytecode.proxy0 extends com.alibaba.dubbo.common.bytecode.Proxy implements com.alibaba.dubbo.common.bytecode.ClassGenerator$DC, com.alibaba.dubbo.rpc.service.EchoService, com.alibaba.mp.alipay.AlipayInfoAbility {
public static java.lang.reflect.Method[] methods;
private java.lang.reflect.InvocationHandler handler;
public (java.lang.reflect.InvocationHandler arg0){
handler = $1;
}
// 代理的接口方法
public com.alibaba.soa.common.response.SoaResponse getAlipayInfo(GetAlipayInfoRequest arg0){
Object[] args = new Object[1];
args[0] = ($w)$1;
// 调用handler的目标方法,此处是 InvokerInvocationHandler
Object ret = handler.invoke(this, methods[0], args);
return (com.alibaba.soa.common.response.SoaResponse)ret;
}
// 默认都会实现 EchoSerice接口,用于回声测试
public java.lang.Object $echo(java.lang.Object arg0) {
Object[] args = new Object[1];
args[0] = ($w)$1;
Object ret = handler.invoke(this, methods[3], args);
return (java.lang.Object)ret;
}
// 创建代理对象
public Object newInstance(java.lang.reflect.InvocationHandler h) {
return new com.alibaba.dubbo.common.bytecode.proxy0($1);
}
③ 看了上述动态生成代理类的过程,再来看 Proxy.getProxy(interfaces).newInstance(new InvokerInvocationHandler(invoker))
这个操作就会很清晰了。拿上面的实现方法来看,当调用ReferenceConcig.get()
获取的对象就是Proxy0
。调用接口的getAlipayInfo
方法,实际上调用的是 Proxy0
的 getAlipayInfo
方法,此代理中,会调用handler的 invoke方法,再来看一下 InvokerInvocationHandler
中的实现逻辑。
④ 正常情况下,我们使用JDK动态代理的时候才会实现java.lang.reflect.InvocationHandler
,此处完全可以不实现InvocationHandler
。但是由于代理的本质是相同的,作者为了代码逻辑复用,统一处理了。(PS: 最后这句是我猜的
)
public class InvokerInvocationHandler implements InvocationHandler {
private final Invoker> invoker; // RegistryAwareClusterInvoker
public InvokerInvocationHandler(Invoker> handler) {
this.invoker = handler;
}
public Object invoke(Object proxy, Method method, Object[] args) throws Throwable {
String methodName = method.getName();
Class>[] parameterTypes = method.getParameterTypes();
if (method.getDeclaringClass() == Object.class) {
return method.invoke(invoker, args);
}
if ("toString".equals(methodName) && parameterTypes.length == 0) {
return invoker.toString();
}
if ("hashCode".equals(methodName) && parameterTypes.length == 0) {
return invoker.hashCode();
}
if ("equals".equals(methodName) && parameterTypes.length == 1) {
return invoker.equals(args[0]);
}
// 此处终于开始调用了。。。
return invoker.invoke(new RpcInvocation(method, args)).recreate();
}
}
- 到此,客户端的初始化Invoker过程已经完毕,后面就是真正开始调用的过程,下一篇我们来分析客户端的调用过程与Netty底层源码解读。
- ☛ 文章要是勘误或者知识点说的不正确,欢迎评论与交流, 希望学习的道路不孤独。由于文章是作者通过阅读源码获得的知识,难免会有疏忽!
- ☛ 要是感觉文章对你有所帮助,不妨点个关注,或者移驾看一下作者的其他文集,也都是干活多多哦,文章也在全力更新中。
- ☛ 著作权归作者所有。商业转载请联系作者获得授权,非商业转载请注明出处!