springCloud-Eureka源码分析
Eureka架构图
Eureka作为springCloud的注册中心,提供了服务注册、服务续约、服务同步等功能,本片文章结合源码来看下Eureka核心功能,原文地址
Eureka核心功能
- 服务注册(register):Eureka Client会通过发送REST请求的方式向Eureka Server注册自己的服务,提供自身的元数
据,比如ip地址、端口、运行状况指标的url、主页地址等信息。Eureka Server接收到注册请求后,就会把这些元数
据信息存储在一个双层的Map中。 - 服务续约(renew):在服务注册后,Eureka Client会维护一个心跳来持续通知Eureka Server,说明服务一直处于可
用状态,防止被剔除。Eureka Client在默认的情况下会每隔30秒(eureka.instance.leaseRenewallIntervalInSeconds)发送一次心跳来进行服务续约。 - 服务同步(replicate):Eureka Server之间会互相进行注册,构建Eureka Server集群,不同Eureka Server之间会进
行服务同步,用来保证服务信息的一致性。 - 获取服务(get registry):服务消费者(Eureka Client)在启动的时候,会发送一个REST请求给Eureka Server,获
取上面注册的服务清单,并且缓存在Eureka Client本地,默认缓存30秒
(eureka.client.registryFetchIntervalSeconds)。同时,为了性能考虑,Eureka Server也会维护一份只读的服务清
单缓存,该缓存每隔30秒更新一次。 - 服务调用:服务消费者在获取到服务清单后,就可以根据清单中的服务列表信息,查找到其他服务的地址,从而进行
远程调用。Eureka有Region和Zone的概念,一个Region可以包含多个Zone,在进行服务调用时,优先访问处于同
一个Zone中的服务提供者。 - 服务下线(cancel):当Eureka Client需要关闭或重启时,就不希望在这个时间段内再有请求进来,所以,就需要提前
先发送REST请求给Eureka Server,告诉Eureka Server自己要下线了,Eureka Server在收到请求后,就会把该服务
状态置为下线(DOWN),并把该下线事件传播出去。 - 服务剔除(evict):有时候,服务实例可能会因为网络故障等原因导致不能提供服务,而此时该实例也没有发送请求给
Eureka Server来进行服务下线,所以,还需要有服务剔除的机制。Eureka Server在启动的时候会创建一个定时任
务,每隔一段时间(默认60秒),从当前服务清单中把超时没有续约(默认90秒,
eureka.instance.leaseExpirationDurationInSeconds)的服务剔除。 - 自我保护:既然Eureka Server会定时剔除超时没有续约的服务,那就有可能出现一种场景,网络一段时间内发生了
异常,所有的服务都没能够进行续约,Eureka Server就把所有的服务都剔除了,这样显然不太合理。所以,就有了
自我保护机制,当短时间内,统计续约失败的比例,如果达到一定阈值,则会触发自我保护的机制,在该机制下,
Eureka Server不会剔除任何的微服务,等到正常后,再退出自我保护机制。自我保护开关(eureka.server.enable-
self-preservation: false)
Eureka Server端源码分析
EurekaServerAutoConfiguration
筛选了部分核心代码进行说明
@Configuration(
proxyBeanMethods = false
)
@Import({EurekaServerInitializerConfiguration.class})
@ConditionalOnBean({Marker.class})
@EnableConfigurationProperties({EurekaDashboardProperties.class, InstanceRegistryProperties.class})
@PropertySource({"classpath:/eureka/server.properties"})
public class EurekaServerAutoConfiguration implements WebMvcConfigurer {
private static final String[] EUREKA_PACKAGES = new String[]{"com.netflix.discovery", "com.netflix.eureka"};
@Autowired
private ApplicationInfoManager applicationInfoManager;
@Autowired
private EurekaServerConfig eurekaServerConfig;
@Autowired
private EurekaClientConfig eurekaClientConfig;
@Autowired
private EurekaClient eurekaClient;
@Autowired
private InstanceRegistryProperties instanceRegistryProperties;
public static final CloudJacksonJson JACKSON_JSON = new CloudJacksonJson();
public EurekaServerAutoConfiguration() {
}
@Bean
public HasFeatures eurekaServerFeature() {
return HasFeatures.namedFeature("Eureka Server", EurekaServerAutoConfiguration.class);
}
// 加载EurekaController, spring‐cloud 提供了一些额外的接口,用来获取eurekaServer的信息
@Bean
@ConditionalOnProperty(
prefix = "eureka.dashboard",
name = {"enabled"},
matchIfMissing = true
)
public EurekaController eurekaController() {
return new EurekaController(this.applicationInfoManager);
}
@Bean
public ServerCodecs serverCodecs() {
return new EurekaServerAutoConfiguration.CloudServerCodecs(this.eurekaServerConfig);
}
private static CodecWrapper getFullJson(EurekaServerConfig serverConfig) {
CodecWrapper codec = CodecWrappers.getCodec(serverConfig.getJsonCodecName());
return codec == null ? CodecWrappers.getCodec(JACKSON_JSON.codecName()) : codec;
}
private static CodecWrapper getFullXml(EurekaServerConfig serverConfig) {
CodecWrapper codec = CodecWrappers.getCodec(serverConfig.getXmlCodecName());
return codec == null ? CodecWrappers.getCodec(XStreamXml.class) : codec;
}
@Bean
@ConditionalOnMissingBean
public ReplicationClientAdditionalFilters replicationClientAdditionalFilters() {
return new ReplicationClientAdditionalFilters(Collections.emptySet());
}
//初始化集群注册表
@Bean
public PeerAwareInstanceRegistry peerAwareInstanceRegistry(ServerCodecs serverCodecs) {
this.eurekaClient.getApplications();
return new InstanceRegistry(this.eurekaServerConfig, this.eurekaClientConfig, serverCodecs, this.eurekaClient, this.instanceRegistryProperties.getExpectedNumberOfClientsSendingRenews(), this.instanceRegistryProperties.getDefaultOpenForTrafficCount());
}
// 配置服务节点信息,这里的作用主要是为了配置Eureka的peer节点,也就是说当有收到有节点注册上来
//的时候,需要通知给那些服务节点, (互为一个集群)
@Bean
@ConditionalOnMissingBean
public PeerEurekaNodes peerEurekaNodes(PeerAwareInstanceRegistry registry, ServerCodecs serverCodecs, ReplicationClientAdditionalFilters replicationClientAdditionalFilters) {
return new EurekaServerAutoConfiguration.RefreshablePeerEurekaNodes(registry, this.eurekaServerConfig, this.eurekaClientConfig, serverCodecs, this.applicationInfoManager, replicationClientAdditionalFilters);
}
//EurekaServer上下文
@Bean
@ConditionalOnMissingBean
public EurekaServerContext eurekaServerContext(ServerCodecs serverCodecs, PeerAwareInstanceRegistry registry, PeerEurekaNodes peerEurekaNodes) {
return new DefaultEurekaServerContext(this.eurekaServerConfig, serverCodecs, registry, peerEurekaNodes, this.applicationInfoManager);
}
// 这个类的作用是spring‐cloud和原生eureka的胶水代码,通过这个类来启动EurekaSever
// 后面这个类会在EurekaServerInitializerConfiguration被调用,进行eureka启动
@Bean
public EurekaServerBootstrap eurekaServerBootstrap(PeerAwareInstanceRegistry registry, EurekaServerContext serverContext) {
return new EurekaServerBootstrap(this.applicationInfoManager, this.eurekaClientConfig, this.eurekaServerConfig, registry, serverContext);
}
// 配置拦截器,ServletContainer里面实现了jersey框架,通过他来实现eurekaServer对外的restFull接口
@Bean
public FilterRegistrationBean<?> jerseyFilterRegistration(Application eurekaJerseyApp) {
FilterRegistrationBean<Filter> bean = new FilterRegistrationBean();
bean.setFilter(new ServletContainer(eurekaJerseyApp));
bean.setOrder(2147483647);
bean.setUrlPatterns(Collections.singletonList("/eureka/*"));
return bean;
}
}
EurekaServerInitializerConfiguration
EurekaServerAutoConfiguration会导入EurekaServerInitializerConfiguration
@Configuration(
proxyBeanMethods = false
)
public class EurekaServerInitializerConfiguration implements ServletContextAware, SmartLifecycle, Ordered {
private static final Log log = LogFactory.getLog(EurekaServerInitializerConfiguration.class);
@Autowired
private EurekaServerConfig eurekaServerConfig;
private ServletContext servletContext;
@Autowired
private ApplicationContext applicationContext;
@Autowired
private EurekaServerBootstrap eurekaServerBootstrap;
private boolean running;
private int order = 1;
public EurekaServerInitializerConfiguration() {
}
public void setServletContext(ServletContext servletContext) {
this.servletContext = servletContext;
}
//启动一个线程
public void start() {
(new Thread(() -> {
try {
//初始化EurekaServer,同时注册Eureka Server
this.eurekaServerBootstrap.contextInitialized(this.servletContext);
log.info("Started Eureka Server");
//发布EurekaServer注册事件
this.publish(new EurekaRegistryAvailableEvent(this.getEurekaServerConfig()));
// 设置启动的状态为true
this.running = true;
// 发送Eureka Start 事件 , 其他还有各种事件,我们可以监听这种时间,然后做一些特定的业务需求
this.publish(new EurekaServerStartedEvent(this.getEurekaServerConfig()));
} catch (Exception var2) {
log.error("Could not initialize Eureka servlet context", var2);
}
})).start();
}
private EurekaServerConfig getEurekaServerConfig() {
return this.eurekaServerConfig;
}
private void publish(ApplicationEvent event) {
this.applicationContext.publishEvent(event);
}
public void stop() {
this.running = false;
this.eurekaServerBootstrap.contextDestroyed(this.servletContext);
}
public boolean isRunning() {
return this.running;
}
public int getPhase() {
return 0;
}
public boolean isAutoStartup() {
return true;
}
public void stop(Runnable callback) {
callback.run();
}
public int getOrder() {
return this.order;
}
}
EurekaServerBootstrap
EurekaServerBootstrap的contextInitialized初始化方法
public class EurekaServerBootstrap {
//初始化EurekaServer的运行环境和上下文
public void contextInitialized(ServletContext context) {
try {
this.initEurekaEnvironment();
this.initEurekaServerContext();
context.setAttribute(EurekaServerContext.class.getName(), this.serverContext);
} catch (Throwable var3) {
log.error("Cannot bootstrap eureka server :", var3);
throw new RuntimeException("Cannot bootstrap eureka server :", var3);
}
}
//初始化EurekaServer的上下文
protected void initEurekaServerContext() throws Exception {
JsonXStream.getInstance().registerConverter(new V1AwareInstanceInfoConverter(), 10000);
XmlXStream.getInstance().registerConverter(new V1AwareInstanceInfoConverter(), 10000);
if (this.isAws(this.applicationInfoManager.getInfo())) {
this.awsBinder = new AwsBinderDelegate(this.eurekaServerConfig, this.eurekaClientConfig, this.registry, this.applicationInfoManager);
this.awsBinder.start();
}
//初始化eureka server上下文
EurekaServerContextHolder.initialize(this.serverContext);
log.info("Initialized server context");
// 从相邻的eureka节点复制注册表
int registryCount = this.registry.syncUp();
// 默认每30秒发送心跳,1分钟就是2次 38
// 修改eureka状态为up 39
// 同时,这里面会开启一个定时任务,用于清理60秒没有心跳的客户端。自动下线
this.registry.openForTraffic(this.applicationInfoManager, registryCount);
EurekaMonitors.registerAllStats();
}
protected void destroyEurekaServerContext() throws Exception {
EurekaMonitors.shutdown();
if (this.awsBinder != null) {
this.awsBinder.shutdown();
}
if (this.serverContext != null) {
this.serverContext.shutdown();
}
}
protected void destroyEurekaEnvironment() throws Exception {
}
protected boolean isAws(InstanceInfo selfInstanceInfo) {
boolean result = Name.Amazon == selfInstanceInfo.getDataCenterInfo().getName();
log.info("isAws returned " + result);
return result;
}
}
public int syncUp() {
int count = 0;
for(int i = 0; i < this.serverConfig.getRegistrySyncRetries() && count == 0; ++i) {
if (i > 0) {
try {
Thread.sleep(this.serverConfig.getRegistrySyncRetryWaitMs());
} catch (InterruptedException var10) {
logger.warn("Interrupted during registry transfer..");
break;
}
}
Applications apps = this.eurekaClient.getApplications();
Iterator var4 = apps.getRegisteredApplications().iterator();
while(var4.hasNext()) {
Application app = (Application)var4.next();
Iterator var6 = app.getInstances().iterator();
while(var6.hasNext()) {
InstanceInfo instance = (InstanceInfo)var6.next();
try {
if (this.isRegisterable(instance)) {
//将其他节点的实例注册到本节点
this.register(instance, instance.getLeaseInfo().getDurationInSecs(), true);
++count;
}
} catch (Throwable var9) {
logger.error("During DS init copy", var9);
}
}
}
}
return count;
}
public void openForTraffic(ApplicationInfoManager applicationInfoManager, int count) {
// 计算每分钟最大续约数
this.expectedNumberOfClientsSendingRenews = count;
this.updateRenewsPerMinThreshold();
logger.info("Got {} instances from neighboring DS node", count);
logger.info("Renew threshold is: {}", this.numberOfRenewsPerMinThreshold);
this.startupTime = System.currentTimeMillis();
if (count > 0) {
this.peerInstancesTransferEmptyOnStartup = false;
}
Name selfName = applicationInfoManager.getInfo().getDataCenterInfo().getName();
boolean isAws = Name.Amazon == selfName;
if (isAws && this.serverConfig.shouldPrimeAwsReplicaConnections()) {
logger.info("Priming AWS connections for all replicas..");
this.primeAwsReplicas(applicationInfoManager);
}
logger.info("Changing status to UP");
//设置实例状态为up
applicationInfoManager.setInstanceStatus(InstanceStatus.UP);
// 开启定时任务,默认60秒执行一次,用于清理60秒之内没有续约的实例
super.postInit();
}
从上面的EurekaServerAutoConfiguration类,我们可以看到有个初始化EurekaServerContext的方法
@Bean
@ConditionalOnMissingBean
public EurekaServerContext eurekaServerContext(ServerCodecs serverCodecs, PeerAwareInstanceRegistry registry, PeerEurekaNodes peerEurekaNodes) {
return new DefaultEurekaServerContext(this.eurekaServerConfig, serverCodecs, registry, peerEurekaNodes, this.applicationInfoManager);
}
DefaultEurekaServerContext 这个类里面的的initialize()方法是被@PostConstruct 这个注解修饰的,
在应用加载的时候,会执行这个方法
@PostConstruct
public void initialize() {
logger.info("Initializing ...");
// 启动一个线程,读取其他集群节点的信息,后面后续复制
this.peerEurekaNodes.start();
try {
this.registry.init(this.peerEurekaNodes);
} catch (Exception var2) {
throw new RuntimeException(var2);
}
logger.info("Initialized");
}
peerEurekaNodes.start()主要是启动一个只拥有一个线程的线程池,第一次进去会更新一下集群其他节点信息 然后启动了一个定时线程,每60秒更新一次,也就是说后续可以根据配置动态的修改节点配置。(原生的spring cloud config支持)
public void start() {
this.taskExecutor = Executors.newSingleThreadScheduledExecutor(new ThreadFactory() {
public Thread newThread(Runnable r) {
Thread thread = new Thread(r, "Eureka-PeerNodesUpdater");
thread.setDaemon(true);
return thread;
}
});
try {
// 首次进来,更新集群节点信息
this.updatePeerEurekaNodes(this.resolvePeerUrls());
//搞个线程
Runnable peersUpdateTask = new Runnable() {
public void run() {
try {
PeerEurekaNodes.this.updatePeerEurekaNodes(PeerEurekaNodes.this.resolvePeerUrls());
} catch (Throwable var2) {
PeerEurekaNodes.logger.error("Cannot update the replica Nodes", var2);
}
}
};
this.taskExecutor.scheduleWithFixedDelay(peersUpdateTask, (long)this.serverConfig.getPeerEurekaNodesUpdateIntervalMs(), (long)this.serverConfig.getPeerEurekaNodesUpdateIntervalMs(), TimeUnit.MILLISECONDS);
} catch (Exception var3) {
throw new IllegalStateException(var3);
}
Iterator var4 = this.peerEurekaNodes.iterator();
while(var4.hasNext()) {
PeerEurekaNode node = (PeerEurekaNode)var4.next();
logger.info("Replica node URL: {}", node.getServiceUrl());
}
}
// 根据URL 构建PeerEurekaNode信息
protected PeerEurekaNode createPeerEurekaNode(String peerEurekaNodeUrl) {
HttpReplicationClient replicationClient = JerseyReplicationClient.createReplicationClient(this.serverConfig, this.serverCodecs, peerEurekaNodeUrl);
String targetHost = hostFromUrl(peerEurekaNodeUrl);
if (targetHost == null) {
targetHost = "host";
}
return new PeerEurekaNode(this.registry, targetHost, peerEurekaNodeUrl, replicationClient, this.serverConfig);
}