- SpringCloud 版本 :Hoxton.SR1
- SpringBoot 版本:2.2.1.RELEASE
- 本文适用于对SpringBoot有一定基础得人,主要讲解Eureka 服务端得相关底层实现,讲解方式:
场景驱动
- 关键词 :服务端源码解析
SpringCloud 服务端与客户端简单搭建 Eureka 服务端与客户端得简单使用
SpringCloud 注解之@EnableEurekaServer与@EnableEurekaClient原理 我们对这两个注解的使用及原理做了分析。
本篇与接下来的连载将会基于服务注册于发现源码层面全面解析
1. Eureka Server
首先 spring-cloud-starter-netflix-eureka-server 依赖下的spring.factories中只有一个自动装配类:
-
类的相关注解部分 :
image.png
- @Import(EurekaServerInitializerConfiguration.class) : 自动装配该类时会创建 EurekaServerInitializerConfiguration的bean,看名字作用是服务初始化的相关操作。另外该类实现了ServletContextAware与SmartLifecycle接口:
image.png ,其中一个重要得地方是生命周期得start()启动方法: image.png ,看到这行代码eurekaServerBootstrap.contextInitialized(EurekaServerInitializerConfiguration.this.servletContext)
,是不是感觉似曾相识,没错就是 ServletContextListener接口得default方法,然后我们猜测这个应该是跟 tomcat容器中运行得servlet组件有关系,但是我们都知道boot依赖得是内置得tomcat容器,而且容器得相关操作是在boot得上下文中进行得,可以看一下我之前得一篇文章 SpringBoot启动 源码深度解析(四),放在此地是何意? 仔细搜寻发现有一个叫 EurekaBootStrap得类实现了ServletContextListener接口,此类是在Eureka中提供得,而且EurekaServerBootstrap中得contextInitialized方法实现
与EurekaBootStrap得contextInitialized方法实现几乎如出一辙
,此时我们可以猜到在Eureka上下文以及环境得初始化与传统得spring运行在tomcat中得方式是一样得,而此处boot照搬了Eureka得初始化代码,通过SmartLifecycle拉起Eureka Server。- @ConditionalOnBean(EurekaServerMarkerConfiguration.Marker.class) :只有存在EurekaServerMarkerConfiguration.Marker 类型的bean才会自动装配当前类。
- @EnableConfigurationProperties({ EurekaDashboardProperties.class,InstanceRegistryProperties.class }) : 发现是@EnableXXX开头的,根据作者之前分析的技巧,查看注解源码: image.png
关于@EnableConfigurationProperties的原理,感兴趣的可以看一下作者之前的文章 SpringBoot配置属性绑定源码分析
👍此处可以得知 @EnableConfigurationProperties 可以创建指定的bean定义,但是导入类要带有@ConfigurationProperties注解才可以👍。以当前导入的EurekaDashboardProperties为例: image.png ,配置的属性 eureka.dashboard.path的值会映射到当前类中的path成员变量上,eureka.dashboard.enabled的值会映射到enabled成员变量上。
-
类的成员变量:
成员变量
- ApplicationInfoManager(原生得Eureka类):用来管理实例信息以及Eureka实例配置(
在EurekaClientAutoConfiguration中创建,后面说到客户端的时候会说明
)- EurekaServerConfig(原生得Eureka接口):用来存储Eureka中得服务配置信息,netflix的默认实现类是DefaultEurekaServerConfig,
对应springcloud中得实现为 EurekaServerConfigBean
- EurekaClientConfig(原生得Eureka接口):用来存储客户端中得配置信息,
对应springcloud中得实现为 EurekaClientConfigBean(在EurekaClientAutoConfiguration中创建,后面说到客户端的时候会说明)
- EurekaClient(原生得Eureka接口):定义了客户端发现得一系列接口协议:① 包含获取实例信息得接口 ② 包含元数据获取得接口 ③ 监控检查得相关方法 ④ Eureka客户端配置信息获取 ⑤ ApplicationInfoManager 单例对象获取,
对应springcloud中得实现是 CloudEurekaClient(在EurekaClientAutoConfiguration中创建,后面说到客户端的时候会说明)
- InstanceRegistryProperties:@EnableConfigurationProperties方式创建的实例配置信息
- EurekaServerAutoConfiguration自动装配类 创建Bean
- 👍创建InstanceRegistry 实例Bean( 在注册实例时会考虑集群情况下其它Node相关操作的注册器):
@Bean
public PeerAwareInstanceRegistry peerAwareInstanceRegistry(
ServerCodecs serverCodecs) {
// 强制初始化
this.eurekaClient.getApplications(); // force initialization
// 创建InstanceRegistry(是spring cloud的实现),
// 继承了PeerAwareInstanceRegistryImpl(Eureka得实现),
// PeerAwareInstanceRegistry接口的实现类
return new InstanceRegistry(this.eurekaServerConfig, this.eurekaClientConfig,
serverCodecs, this.eurekaClient,
this.instanceRegistryProperties.getExpectedNumberOfClientsSendingRenews(),
this.instanceRegistryProperties.getDefaultOpenForTrafficCount());
}
- 👍创建 PeerEurekaNodes(用来管理PeerEurekaNode的生命周期辅助类), 实际创建子类 RefreshablePeerEurekaNodes,该类是当前自动装配类得内部静态类,另外该类还实现了ApplicationListener监听器,监听EnvironmentChangeEvent事件, 触发Eureka节点得更新
@Bean
@ConditionalOnMissingBean
public PeerEurekaNodes peerEurekaNodes(PeerAwareInstanceRegistry registry,
ServerCodecs serverCodecs,
ReplicationClientAdditionalFilters replicationClientAdditionalFilters) {
return new RefreshablePeerEurekaNodes(registry, this.eurekaServerConfig,
this.eurekaClientConfig, serverCodecs, this.applicationInfoManager,
replicationClientAdditionalFilters);
}
子类RefreshablePeerEurekaNodes中得方法
父类PeerEurekaNodes中得方法
① createPeerEurekaNode方法 :覆写了父类 PeerEurekaNodes,与父类中得实现唯一区别是添加了一个自定义过滤器 ReplicationClientAdditionalFilters。
② shouldUpdate方法:当监听到EnvironmentChangeEvent改变事件时,判断是否需要更新,依据是环境资源中是否包含 eureka.client.region、eureka.client.service-url、eureka.client.availability-zones. 这些键,若包含则需要更新。
③ resolvePeerUrls()方法:返回不包含自己得可用区的服务端url。
④ updatePeerEurekaNodes方法,继承自父类,首先从原先得Eureka Server 列表中移除新增得List列表,目的是找出不可用得服务列表,再从新增得列表中移除原先得(peerEurekaNodeUrls 变量中缓存得),目的是找出实际新增得服务列表,紧接着判断若两个集合是否为空,都为空则不更新;若shutdown列表不为空,找到shutdown得节点,调用PeerEurekaNode(具体得node)得shutdown方法关闭副本资源,若新增得服务列表不为空,则调用createPeerEurekaNode方法(前面提到得得方法,被子类覆写)创建新增得 PeerEurekaNode节点,最终将Eureka Server 节点保存在成员变量 this.peerEurekaNodes中;将Eureka Server 节点url保存到成员变量this.peerEurekaNodeUrls中。
- 👍👍创建 EurekaServerConfig 实例(Eureka中得类),创建动作交给EurekaServerAutoConfiguration 类得内部类 EurekaServerConfigBeanConfiguration 来完成,实现为 EurekaServerConfigBean( springcloud 中 Eureka Server得核心类),包括一系列得服务端参数配置,服务注册与续约等相关功能都会根据此类中得参数配置进行处理,示例: application.properties
@Configuration(proxyBeanMethods = false)
protected static class EurekaServerConfigBeanConfiguration {
@Bean
@ConditionalOnMissingBean
public EurekaServerConfig eurekaServerConfig(EurekaClientConfig clientConfig) {
EurekaServerConfigBean server = new EurekaServerConfigBean();
if (clientConfig.shouldRegisterWithEureka()) {
// Set a sensible default if we are supposed to replicate
server.setRegistrySyncRetries(5);
}
return server;
}
}
我们发现registerWithEureka参数是从 EurekaClientConfigBean(客户端参数配置类,服务端是一个特殊得客户端,在EurekaClientAutoConfiguration中创建,后面说到客户端的时候会说明
) 中获取得,若为false,那么重试次数就为0,默认该参数为 true,重试次数为5
- 👍👍创建 EurekaServerContext,Eureka Server上下文,实际创建得是默认实现 DefaultEurekaServerContext类,可以看到将上面得众多参数构造器注入进去,其中就有 PeerEurekaNodes和前面创建得相关Bean实例,此处我们重点分析。
@Bean
public EurekaServerContext eurekaServerContext(ServerCodecs serverCodecs,
PeerAwareInstanceRegistry registry, PeerEurekaNodes peerEurekaNodes) {
return new DefaultEurekaServerContext(this.eurekaServerConfig, serverCodecs,
registry, peerEurekaNodes, this.applicationInfoManager);
}
进入到 DefaultEurekaServerContext 中,源码如下:
/**
* Represent the local server context and exposes getters to components of the
* local server such as the registry.
*
* @author David Liu
*/
@Singleton
public class DefaultEurekaServerContext implements EurekaServerContext {
private static final Logger logger = LoggerFactory.getLogger(DefaultEurekaServerContext.class);
private final EurekaServerConfig serverConfig;
private final ServerCodecs serverCodecs;
private final PeerAwareInstanceRegistry registry;
private final PeerEurekaNodes peerEurekaNodes;
private final ApplicationInfoManager applicationInfoManager;
@Inject
public DefaultEurekaServerContext(EurekaServerConfig serverConfig,
ServerCodecs serverCodecs,
PeerAwareInstanceRegistry registry,
PeerEurekaNodes peerEurekaNodes,
ApplicationInfoManager applicationInfoManager) {
this.serverConfig = serverConfig;
this.serverCodecs = serverCodecs;
this.registry = registry;
this.peerEurekaNodes = peerEurekaNodes;
this.applicationInfoManager = applicationInfoManager;
}
@PostConstruct
@Override
public void initialize() {
logger.info("Initializing ...");
peerEurekaNodes.start();
try {
registry.init(peerEurekaNodes);
} catch (Exception e) {
throw new RuntimeException(e);
}
logger.info("Initialized");
}
@PreDestroy
@Override
public void shutdown() {
logger.info("Shutting down ...");
registry.shutdown();
peerEurekaNodes.shutdown();
logger.info("Shut down");
}
@Override
public EurekaServerConfig getServerConfig() {return serverConfig;}
@Override
public PeerEurekaNodes getPeerEurekaNodes() {return peerEurekaNodes;}
@Override
public ServerCodecs getServerCodecs() {return serverCodecs;}
@Override
public PeerAwareInstanceRegistry getRegistry() { return registry;}
@Override
public ApplicationInfoManager getApplicationInfoManager() {return applicationInfoManager;}
}
① initialize()方法:可以看到使用@PostConstruct修饰,表示实例化之后初始化之前 Spring Bean 创建得生命周期,会调用此方法,(1)首先调用 Eureka Server 节点得start()方法:
peerEurekaNodes.start() , 方法中会创建单个线程得线程池,线程名叫做Eureka-PeerNodesUpdater,然后调用上述提到得 updatePeerEurekaNodes(List<String> newPeerUrls)方法 立即更新一次Eureka Server得节点,然后每隔10分钟定时更新服务列表。(前面还说过监听到事件也可能会更新服务列表)(2)然后通过实例注册器(InstanceRegistry,继承了 PeerAwareInstanceRegistryImpl)调用 registry.init(peerEurekaNodes) 初始化服务节点: PeerAwareInstanceRegistryImpl中得init方法this.numberOfReplicationsLastMin.start()会启动一个Timer定时器每隔一分钟记录复制的续约数;赋值成员变量 peerEurekaNodes;👍initializedResponseCache()方法会初始化ResponseCache(缓存得是服务列表) :
// 只读的服务列表
private final ConcurrentMap<Key, Value> readOnlyCacheMap = new ConcurrentHashMap<Key, Value>();
// Guava 缓存,读写服务列表
private final LoadingCache<Key, Value> readWriteCacheMap;
ResponseCacheImpl(EurekaServerConfig serverConfig, ServerCodecs serverCodecs, AbstractInstanceRegistry registry) {
this.serverConfig = serverConfig;
this.serverCodecs = serverCodecs;
// 根据配置eureka.server.useReadOnlyResponseCache判断,是否使用只读ResponseCache,默认true
// 由于ResponseCache维护这一个可读可写的readWriteCacheMap,还有一个只读的readOnlyCacheMap
// 此配置控制在get()应用数据时,是去只读Map读,还是读写Map读
this.shouldUseReadOnlyResponseCache = serverConfig.shouldUseReadOnlyResponseCache();
this.registry = registry;
// eureka.server.responseCacheUpdateIntervalMs缓存更新频率,默认30s
long responseCacheUpdateIntervalMs = serverConfig.getResponseCacheUpdateIntervalMs();
// 创建Cache,com.google.common.cache.LoadingCache
// 可以设置初始值(默认1000),数据写入过期时间(默认180秒),过期清理等
this.readWriteCacheMap = CacheBuilder.newBuilder()
.initialCapacity(serverConfig.getInitialCapacityOfResponseCache())
.expireAfterWrite(serverConfig.getResponseCacheAutoExpirationInSeconds(), TimeUnit.SECONDS)
.removalListener(new RemovalListener<Key, Value>() {
@Override
public void onRemoval(RemovalNotification<Key, Value> notification) {
Key removedKey = notification.getKey();
if (removedKey.hasRegions()) {
Key cloneWithNoRegions = removedKey.cloneWithoutRegions();
regionSpecificKeys.remove(cloneWithNoRegions, removedKey);
}
}
})
.build(new CacheLoader<Key, Value>() {
@Override
public Value load(Key key) throws Exception {
if (key.hasRegions()) {
Key cloneWithNoRegions = key.cloneWithoutRegions();
regionSpecificKeys.put(cloneWithNoRegions, key);
}
Value value = generatePayload(key);
return value;
}
});
// 如果启用只读缓存(默认开启),那么每隔responseCacheUpdateIntervalMs=30s,执行缓存更新
// (从两个缓存中取出值进行对比,若不相等则更新只读缓存)
if (shouldUseReadOnlyResponseCache) {
timer.schedule(getCacheUpdateTask(),
new Date(((System.currentTimeMillis() / responseCacheUpdateIntervalMs) * responseCacheUpdateIntervalMs)
+ responseCacheUpdateIntervalMs),
responseCacheUpdateIntervalMs);
}
try {
Monitors.registerObject(this);
} catch (Throwable e) {
logger.warn("Cannot register the JMX monitor for the InstanceRegistry", e);
}
}
ResponseCache 得作用是保存服务列表信息,用于客户端查询。实现上内部维护了两个Map,一个可读可写的readWriteCacheMap,每个操作都会写入,一个只读的readOnlyCacheMap,默认每30s更新一次, 其中缓存以压缩和非压缩形式维护,用于三类请求: all applications,增量更改和单个application,👍例如在ApplicationResource(Eureka暴漏得服务实例接口)中:
ApplicationResource.getApplication方法 此处查询时,首先去缓存中查找,若 this.shouldUseReadOnlyResponseCache = serverConfig.shouldUseReadOnlyResponseCache(),默认为true,表示查询的时候从readOnlyCacheMap中查找,否则从readWriteCacheMap中查找,而readWriteCacheMap中得数据会每隔30秒同步到readOnlyCacheMap中。然后成员变量 timer,名称为 ReplicaAwareInstanceRegistry - RenewalThresholdUpdater
得Time定时器 每隔15分钟调用 updateRenewalThreshold方法 更新续约临界值。
private void scheduleRenewalThresholdUpdateTask() {
timer.schedule(new TimerTask() {
@Override
public void run() {
updateRenewalThreshold(); // 更新续约临界值
}
}, serverConfig.getRenewalThresholdUpdateIntervalMs(),// 15 min
serverConfig.getRenewalThresholdUpdateIntervalMs()); // 15 min
}
private void updateRenewalThreshold() {
try {
Applications apps = eurekaClient.getApplications();
int count = 0;
for (Application app : apps.getRegisteredApplications()) {
for (InstanceInfo instance : app.getInstances()) {
if (this.isRegisterable(instance)) {
++count; // 记录实例总数
}
}
}
synchronized (lock) {
//(1)只有当临界值比 (期望续约得数量 乘以 续约临界值百分比默认 0.85)大,才更新,即:服务实例数大于85%才会去更新,小于85%不去更新续约服务
//(2)不启用自我保护机制(默认开启)
if ((count) > (serverConfig.getRenewalPercentThreshold() * expectedNumberOfClientsSendingRenews)
|| (!this.isSelfPreservationModeEnabled())) {
this.expectedNumberOfClientsSendingRenews = count;
updateRenewsPerMinThreshold();
}
}
logger.info("Current renewal threshold is : {}", numberOfRenewsPerMinThreshold);
} catch (Throwable e) {
logger.error("Cannot update renewal threshold", e);
}
}
首先获取所有Instance实例个数,默认每个实例30s续约一次(1)如果开启自我保护模式,更新 expectedNumberOfRenewsPerMin预期每分钟续约数 和 numberOfRenewsPerMinThreshold每分钟续约阈值 (2)如果没有开启自我保护模式,只有当本期续约数大于之前的阈值,即当前不处在自我保护模式中(自我保护模式中,不能删除服务列表,阈值自然也不能更新),才可以更新 expectedNumberOfRenewsPerMin 和 numberOfRenewsPerMinThreshold。
最后调用initRemoteRegionRegistry()方法初始化远程Region注册器(若当前Region对应得实例宕机,可以拉取其他Region下面得实例进行调用)
② shutdown方法:标有 @PreDestroy注解,Bean销毁之前调用,用以关闭Eureka Server 节点以及一系统得清除重置操作。
- 创建 EurekaServerBootstrap( springcloud得类 ):
@Bean
public EurekaServerBootstrap eurekaServerBootstrap(PeerAwareInstanceRegistry registry,
EurekaServerContext serverContext) {
return new EurekaServerBootstrap(this.applicationInfoManager,
this.eurekaClientConfig, this.eurekaServerConfig, registry,
serverContext);
}
上面已经大致分析了EurekaServerBootstrap类得作用,此处中重点分析一些类中得方法实现细节:
EurekaServerBootstrap① contextInitialized(ServletContext context)方法,spring容器启动之后会回调生命周期中得start启动方法,拉起 contextInitialized方法初始化Eureka Server。首先调用 initEurekaEnvironment方法 初始化Eureka Server得环境 ( 此处配置就是Netflix Archaius,都是基于Apache Commons 抽象 Configuration来实现得),然后调用initEurekaServerContext 方法初始化Eureka Server上下文,并将上下文缓存到 EurekaServerContextHolder中。 image.png
有两处重要的逻辑
:(1) int registryCount = this.registry.syncUp() 表示从相邻的eureka节点拷贝注册列表信息
@Override
public int syncUp() {
// Copy entire entry from neighboring DS node
int count = 0;
// 此处重试次数为5
for (int i = 0; ((i < serverConfig.getRegistrySyncRetries()) && (count == 0)); i++) {
if (i > 0) {
try {
Thread.sleep(serverConfig.getRegistrySyncRetryWaitMs());
} catch (InterruptedException e) {
logger.warn("Interrupted during registry transfer..");
break;
}
}
// 通过 eurekaClient(根据上面的简介可以知道此处是CloudEurekaClient实现) 获取全部服务
Applications apps = eurekaClient.getApplications();
for (Application app : apps.getRegisteredApplications()) {
// 循环获取实例列表(注册的时候多个服务实例可以用同一个应用名)
for (InstanceInfo instance : app.getInstances()) {
try {
// 判断是否是可以注册得:正常情况下此方法一定返回 true
if (isRegisterable(instance)) {
// 注册实例 ( 注册逻辑得具体实现,在AbstractInstanceRegistry类中
register(instance, instance.getLeaseInfo().getDurationInSecs(), true);
// 统计注册数(后续续约的时候会用到)
count++;
}
} catch (Throwable t) {
logger.error("During DS init copy", t);
}
}
}
}
return count;
}
判断如果可注册,然后接着执行com.netflix.eureka.registry.AbstractInstanceRegistry#register方法执行具体得注册逻辑,可以看出此方法是在抽象父类AbstractInstanceRegistry中,但是InstanceRegistry(springcloud得bean,并且继承了PeerAwareInstanceRegistryImpl类)和PeerAwareInstanceRegistryImpl(Eureka中得注册实例对实现)都覆写了该方法:
InstanceRegistry得register方法实现 PeerAwareInstanceRegistryImpl中得registry方法依次分析,可以看到此处handleRegistration方法中发布了一个EurekaInstanceRegisteredEvent事件,此事件对象会包含InstanceInfo(注册得当前实例信息)、leaseDuration(租约期限)、isReplication(是否是复制来得),
另外事件源source就是当前InstanceRegistry对象本身。所以,我们这里可以进行扩展(比如:我们可以监听该事件,从而获得当前注册实例,进行自定义的逻辑处理)。
然后调用父类PeerAwareInstanceRegistryImpl得register方法,红框中得方法重点分析
首先判断若当前实例中若没有租约时间则使用默认得租约时间90秒,然后再调用抽象父类AbstractInstanceRegistry得register方法 :
/**
* Registers a new instance with a given duration.
*
* @see com.netflix.eureka.lease.LeaseManager#register(java.lang.Object, int, boolean)
*/
public void register(InstanceInfo registrant, int leaseDuration, boolean isReplication) {
try {
// 获取并发读写锁。
read.lock();
// 从缓存中,根据应用名称查看是否包含当前实例得缓存注册信息
Map<String, Lease<InstanceInfo>> gMap = registry.get(registrant.getAppName());
// 统计计数器+1
REGISTER.increment(isReplication);
// 若缓存中不存在当前应用实例信息,创建一个空得CurrentHashMap对象,向缓存中添加这个CurrentHashMap对象
if (gMap == null) {
final ConcurrentHashMap<String, Lease<InstanceInfo>> gNewMap = new ConcurrentHashMap<String, Lease<InstanceInfo>>();
gMap = registry.putIfAbsent(registrant.getAppName(), gNewMap);
if (gMap == null) {
gMap = gNewMap;
}
}
// 获取实例的租约信息
Lease<InstanceInfo> existingLease = gMap.get(registrant.getId());
// 表示当前实例已经有过租约了
if (existingLease != null && (existingLease.getHolder() != null)) {
Long existingLastDirtyTimestamp = existingLease.getHolder().getLastDirtyTimestamp();
Long registrationLastDirtyTimestamp = registrant.getLastDirtyTimestamp();
logger.debug("Existing lease found (existing={}, provided={}", existingLastDirtyTimestamp, registrationLastDirtyTimestamp);
// 已经有过租约得时间戳比当前要注册得时间戳大,
// 那么将替换掉要注册得实例为已经有过租约得实例
if (existingLastDirtyTimestamp > registrationLastDirtyTimestamp) {
logger.warn("There is an existing lease and the existing lease's dirty timestamp {} is greater" +
" than the one that is being registered {}", existingLastDirtyTimestamp, registrationLastDirtyTimestamp);
logger.warn("Using the existing instanceInfo instead of the new instanceInfo as the registrant");
registrant = existingLease.getHolder();
}
} else {
// 表示新注册得实例不存在租约
synchronized (lock) {
// 已经有过其他服务实例得租约信息了
if (this.expectedNumberOfClientsSendingRenews > 0) {
// 将旧的客户端要发送得续约数量得数值加1,默认数值是1;
// 即第一次注册的时候,该变量数值为2
this.expectedNumberOfClientsSendingRenews = this.expectedNumberOfClientsSendingRenews + 1;
// 重新计算每分钟续约得临界值,
// 默认情况下是
// this.expectedNumberOfClientsSendingRenews *(60/30) * 0.85
updateRenewsPerMinThreshold();
}
}
logger.debug("No previous lease information found; it is new registration");
}
// 重新创建租约信息
Lease<InstanceInfo> lease = new Lease<InstanceInfo>(registrant, leaseDuration);
// 若旧的租约信息存在,将旧的租约服务启动时间复制给新创建得租约信息
if (existingLease != null) {
lease.setServiceUpTimestamp(existingLease.getServiceUpTimestamp());
}
gMap.put(registrant.getId(), lease);
synchronized (recentRegisteredQueue) {
recentRegisteredQueue.add(new Pair<Long, String>(
System.currentTimeMillis(),
registrant.getAppName() + "(" + registrant.getId() + ")"));
}
// 如果当前实例已经维护了OverriddenStatus,将其也放到此Eureka Server
// 的overriddenInstanceStatusMap中
if (!InstanceStatus.UNKNOWN.equals(registrant.getOverriddenStatus())) {
logger.debug("Found overridden status {} for instance {}. Checking to see if needs to be add to the "
+ "overrides", registrant.getOverriddenStatus(), registrant.getId());
if (!overriddenInstanceStatusMap.containsKey(registrant.getId())) {
logger.info("Not found overridden id {} and hence adding it", registrant.getId());
overriddenInstanceStatusMap.put(registrant.getId(), registrant.getOverriddenStatus());
}
}
InstanceStatus overriddenStatusFromMap = overriddenInstanceStatusMap.get(registrant.getId());
if (overriddenStatusFromMap != null) {
logger.info("Storing overridden status {} from map", overriddenStatusFromMap);
registrant.setOverriddenStatus(overriddenStatusFromMap);
}
// 根据实例得overridden状态规则,设置状态
InstanceStatus overriddenInstanceStatus = getOverriddenInstanceStatus(registrant, existingLease, isReplication);
registrant.setStatusWithoutDirty(overriddenInstanceStatus);
// 如果注册得租约状态为UP(准备接收请求状态),设置租约服务时间戳为当前时间
if (InstanceStatus.UP.equals(registrant.getStatus())) {
lease.serviceUp();
}
registrant.setActionType(ActionType.ADDED);
recentlyChangedQueue.add(new RecentlyChangedItem(lease));
registrant.setLastUpdatedTimestamp();
// 使当前应用的ResponseCache缓存失效
invalidateCache(registrant.getAppName(), registrant.getVIPAddress(), registrant.getSecureVipAddress());
logger.info("Registered instance {}/{} with status {} (replication={})",
registrant.getAppName(), registrant.getId(), registrant.getStatus(), isReplication);
} finally {
read.unlock();
}
}
调用完父类的注册方法之后,在PeerAwareInstanceRegistryImpl中接着执行replicateToPeers方法:将实例信息复制到集群中其它节点
private void replicateToPeers(Action action, String appName, String id,
InstanceInfo info /* optional */,
InstanceStatus newStatus /* optional */, boolean isReplication) {
Stopwatch tracer = action.getTimer().start();
try {
// 若是复制来的实例,则将记录的最后一分钟的副本数自增1
if (isReplication) {
numberOfReplicationsLastMin.increment();
}
// 一个对象为什么跟一个集合用等号判断 ?
// (由于启动时调用的注册传入标志为true,所以会结束流程)
if (peerEurekaNodes == Collections.EMPTY_LIST || isReplication) {
return;
}
// 遍历所有得服务节点
for (final PeerEurekaNode node : peerEurekaNodes.getPeerEurekaNodes()) {
// 跳过自身得服务url
if (peerEurekaNodes.isThisMyUrl(node.getServiceUrl())) {
continue;
}
// 将对应得行为(注册、下线、续约、心跳、状态更新)复制到其他节点
replicateInstanceActionsToPeers(action, appName, id, info, newStatus, node);
}
} finally {
tracer.stop();
}
}
(2) registry.openForTraffic(this.applicationInfoManager, regisretryCount): 允许与客户端的数据传输。其中registryCount为 syncUp方法调用得返回值:可注册的服务实例数(为0表示单机)。通过先调用子类InstanceRegistry,若此时的regisretryCount为0,则取默认值1,然后调用父类的openForTraffic的方法:
@Override
public void openForTraffic(ApplicationInfoManager applicationInfoManager, int count) {
// 设置续约数为注册的数量,默认每30秒续约一次
this.expectedNumberOfClientsSendingRenews = count;
// 更新每分钟续约临界值(前面分析过,就是:续约数 * 每分钟次数(2次) * 临界值百分比(0.85))
updateRenewsPerMinThreshold();
logger.info("Got {} instances from neighboring DS node", count);
logger.info("Renew threshold is: {}", numberOfRenewsPerMinThreshold);
this.startupTime = System.currentTimeMillis();
if (count > 0) {
this.peerInstancesTransferEmptyOnStartup = false;
}
DataCenterInfo.Name selfName = applicationInfoManager.getInfo().getDataCenterInfo().getName();
boolean isAws = Name.Amazon == selfName;
if (isAws && serverConfig.shouldPrimeAwsReplicaConnections()) {
logger.info("Priming AWS connections for all replicas..");
primeAwsReplicas(applicationInfoManager);
}
logger.info("Changing status to UP");
// 设置实例信息状态为UP上线,并向监听器StatusChangeListener的实例发送StatusChangeEvent事件,
// 告知实例状态改变,当然这些监听器取决于clientConfig客户端配置的onDemandUpdateStatusChange
// 参数,为true的话,会添加监听器用于监听状态改变,否则不通知。
applicationInfoManager.setInstanceStatus(InstanceStatus.UP);
super.postInit();
}
- 👍👍PeerEurekaNode( Eureka得原生类,真实的服务端节点,会保存服务端的url、服务配置信息、服务端注册器、目标主机、httpClient、批量任务转发器、非批量任务转发器):前面分析过 PeerEurekaNodes#updatePeerEurekaNodes方法时判断出url属于新增的话,则会创建PeerEurekaNode节点。而PeerEurekaNode构造器里面会初始化一系列的成员:
public PeerEurekaNode(PeerAwareInstanceRegistry registry, String targetHost, String serviceUrl, HttpReplicationClient replicationClient, EurekaServerConfig config) {
// 最大批次请求250个,最大批量任务转发等待时间为500毫秒
// 请求异常情况下会隔100毫秒重试,服务不可用等待1000毫秒
this(registry, targetHost, serviceUrl, replicationClient, config, BATCH_SIZE, MAX_BATCHING_DELAY_MS, RETRY_SLEEP_TIME_MS, SERVER_UNAVAILABLE_SLEEP_TIME_MS);
}
PeerEurekaNode(PeerAwareInstanceRegistry registry, String targetHost, String serviceUrl,
HttpReplicationClient replicationClient, EurekaServerConfig config,
int batchSize, long maxBatchingDelayMs,
long retrySleepTimeMs, long serverUnavailableSleepTimeMs) {
this.registry = registry;
this.targetHost = targetHost;
this.replicationClient = replicationClient;
this.serviceUrl = serviceUrl;
this.config = config;
this.maxProcessingDelayMs = config.getMaxTimeForReplication();
String batcherName = getBatcherName();
// 复制任务处理器
ReplicationTaskProcessor taskProcessor = new ReplicationTaskProcessor(targetHost, replicationClient);
// 批量任务转发器
this.batchingDispatcher = TaskDispatchers.createBatchingTaskDispatcher(
batcherName,
config.getMaxElementsInPeerReplicationPool(),
batchSize,
config.getMaxThreadsForPeerReplication(),
maxBatchingDelayMs,
serverUnavailableSleepTimeMs,
retrySleepTimeMs,
taskProcessor
);
// 单任务转发器
this.nonBatchingDispatcher = TaskDispatchers.createNonBatchingTaskDispatcher(
targetHost,
config.getMaxElementsInStatusReplicationPool(),
config.getMaxThreadsForStatusReplication(),
maxBatchingDelayMs,
serverUnavailableSleepTimeMs,
retrySleepTimeMs,
taskProcessor
);
}
// 任务转发器工具类,用来创建TaskDispatcher,想象成Executors线程池工具类就行
public class TaskDispatchers {
// 创建批量任务转发器
public static <ID, T> TaskDispatcher<ID, T> createBatchingTaskDispatcher(String id,
int maxBufferSize,
int workloadSize,
int workerCount,
long maxBatchingDelay,
long congestionRetryDelayMs,
long networkFailureRetryMs,
TaskProcessor<T> taskProcessor){
// 创建Acceptor执行器,构造器中会创建线程组并分配AcceptorThread
// 然后执行AcceptorRunner任务
final AcceptorExecutor<ID, T> acceptorExecutor = new AcceptorExecutor<>(
id, maxBufferSize, workloadSize, maxBatchingDelay, congestionRetryDelayMs, networkFailureRetryMs
);
// 创建TaskExecutors并启动工作线程WorkerRunnable(后面详细看工作线程如何工作的)
final TaskExecutors<ID, T> taskExecutor = TaskExecutors.batchExecutors(id, workerCount, taskProcessor, acceptorExecutor);
return new TaskDispatcher<ID, T>() {
@Override
// 任务转发器的处理方法,批量任务转发器调用统一入口
public void process(ID id, T task, long expiryTime) {
// 将任务添加到AcceptorExecutor类中的acceptorQueue队列里
// 并将接收任务数+1
acceptorExecutor.process(id, task, expiryTime);
}
@Override
public void shutdown() {
// 将AcceptorExecutor类中的isShutdown状态设置为true
// 并中断AcceptorThread线程
acceptorExecutor.shutdown();
// 将TaskExecutors类中的isShutdown状态也设置为true
// 并循环中断所有的工作线程
taskExecutor.shutdown();
}
};
}
}
// Acceptor线程的执行器
class AcceptorExecutor<ID, T> {
AcceptorExecutor(String id,
int maxBufferSize,
int maxBatchingSize,
long maxBatchingDelay,
long congestionRetryDelayMs,
long networkFailureRetryMs) {
this.maxBufferSize = maxBufferSize;
this.maxBatchingSize = maxBatchingSize;
this.maxBatchingDelay = maxBatchingDelay;
this.trafficShaper = new TrafficShaper(congestionRetryDelayMs, networkFailureRetryMs);
// 创建名为eurekaTaskExecutors的线程组
ThreadGroup threadGroup = new ThreadGroup("eurekaTaskExecutors");
// 创建线程并指定线程组,接收AcceptorRunner任务
// id:target_主机名;所以线程名称:TaskAcceptor-target_主机名
this.acceptorThread = new Thread(threadGroup, new AcceptorRunner(), "TaskAcceptor-" + id);
this.acceptorThread.setDaemon(true);
// 然后启动线程
this.acceptorThread.start();
// 记录指标
final double[] percentiles = {50.0, 95.0, 99.0, 99.5};
final StatsConfig statsConfig = new StatsConfig.Builder()
.withSampleSize(1000)
.withPercentiles(percentiles)
.withPublishStdDev(true)
.build();
final MonitorConfig config = MonitorConfig.builder(METRIC_REPLICATION_PREFIX + "batchSize").build();
this.batchSizeMetric = new StatsTimer(config, statsConfig);
try {
Monitors.registerObject(id, this);
} catch (Throwable e) {
logger.warn("Cannot register servo monitor for this object", e);
}
}
}
// Acceptor的任务类型
class AcceptorRunner implements Runnable {
@Override
public void run() {
long scheduleTime = 0;
// 任务没有关闭
while (!isShutdown.get()) {
try {
// 从队列中取出所有的任务并执行
drainInputQueues();
int totalItems = processingOrder.size();
long now = System.currentTimeMillis();
if (scheduleTime < now) {
scheduleTime = now + trafficShaper.transmissionDelay();
}
if (scheduleTime <= now) {
assignBatchWork();
assignSingleItemWork();
}
// If no worker is requesting data or there is a delay injected by the traffic shaper,
// sleep for some time to avoid tight loop.
if (totalItems == processingOrder.size()) {
Thread.sleep(10);
}
} catch (InterruptedException ex) {
// Ignore
} catch (Throwable e) {
// Safe-guard, so we never exit this loop in an uncontrolled way.
logger.warn("Discovery AcceptorThread error", e);
}
}
}
private boolean isFull() {
return pendingTasks.size() >= maxBufferSize;
}
private void drainInputQueues() throws InterruptedException {
do {
// 从重试队列队尾循环取出任务,判断任务改怎么处理
drainReprocessQueue();
// 当调用PeerEurekaNode中创建的批量任务转发器 batchingDispatcher的
// process方法,会将任务放入到AcceptorExecutor类中的acceptorQueue
// 队列里
// 此处是取出acceptorQueue队列中得任务,处理方式跟上述差不多
drainAcceptorQueue();
if (!isShutdown.get()) {
// If all queues are empty, block for a while on the acceptor queue
if (reprocessQueue.isEmpty() && acceptorQueue.isEmpty() && pendingTasks.isEmpty()) {
TaskHolder<ID, T> taskHolder = acceptorQueue.poll(10, TimeUnit.MILLISECONDS);
if (taskHolder != null) {
appendTaskHolder(taskHolder);
}
}
}
} while (!reprocessQueue.isEmpty() || !acceptorQueue.isEmpty() || pendingTasks.isEmpty());
}
private void drainAcceptorQueue() {
while (!acceptorQueue.isEmpty()) {
appendTaskHolder(acceptorQueue.poll());
}
}
private void drainReprocessQueue() {
long now = System.currentTimeMillis();
while (!reprocessQueue.isEmpty() && !isFull()) {
TaskHolder<ID, T> taskHolder = reprocessQueue.pollLast();
ID id = taskHolder.getId();
if (taskHolder.getExpiryTime() <= now) {
expiredTasks++;
} else if (pendingTasks.containsKey(id)) {
overriddenTasks++;
} else {
pendingTasks.put(id, taskHolder);
processingOrder.addFirst(id);
}
}
if (isFull()) {
queueOverflows += reprocessQueue.size();
reprocessQueue.clear();
}
}
private void appendTaskHolder(TaskHolder<ID, T> taskHolder) {
if (isFull()) {
pendingTasks.remove(processingOrder.poll());
queueOverflows++;
}
TaskHolder<ID, T> previousTask = pendingTasks.put(taskHolder.getId(), taskHolder);
if (previousTask == null) {
processingOrder.add(taskHolder.getId());
} else {
overriddenTasks++;
}
}
void assignSingleItemWork() {
if (!processingOrder.isEmpty()) {
if (singleItemWorkRequests.tryAcquire(1)) {
long now = System.currentTimeMillis();
while (!processingOrder.isEmpty()) {
ID id = processingOrder.poll();
TaskHolder<ID, T> holder = pendingTasks.remove(id);
if (holder.getExpiryTime() > now) {
singleItemWorkQueue.add(holder);
return;
}
expiredTasks++;
}
singleItemWorkRequests.release();
}
}
}
void assignBatchWork() {
if (hasEnoughTasksForNextBatch()) {
if (batchWorkRequests.tryAcquire(1)) {
long now = System.currentTimeMillis();
int len = Math.min(maxBatchingSize, processingOrder.size());
List<TaskHolder<ID, T>> holders = new ArrayList<>(len);
while (holders.size() < len && !processingOrder.isEmpty()) {
ID id = processingOrder.poll();
TaskHolder<ID, T> holder = pendingTasks.remove(id);
if (holder.getExpiryTime() > now) {
holders.add(holder);
} else {
expiredTasks++;
}
}
if (holders.isEmpty()) {
batchWorkRequests.release();
} else {
batchSizeMetric.record(holders.size(), TimeUnit.MILLISECONDS);
batchWorkQueue.add(holders);
}
}
}
}
private boolean hasEnoughTasksForNextBatch() {
if (processingOrder.isEmpty()) {
return false;
}
if (pendingTasks.size() >= maxBufferSize) {
return true;
}
TaskHolder<ID, T> nextHolder = pendingTasks.get(processingOrder.peek());
long delay = System.currentTimeMillis() - nextHolder.getSubmitTimestamp();
return delay >= maxBatchingDelay;
}
}
// 批量任务执行器工具类
class TaskExecutors<ID, T> {
// 下面的方法会最终调用构造器创建批量任务执行器工具类TaskExecutors对象
TaskExecutors(WorkerRunnableFactory<ID, T> workerRunnableFactory, int workerCount, AtomicBoolean isShutdown) {
this.isShutdown = isShutdown;
this.workerThreads = new ArrayList<>();
// 指定线程组名为:eurekaTaskExecutors
ThreadGroup threadGroup = new ThreadGroup("eurekaTaskExecutors");
// 此处的workerCount可在EurekaServerConfig中配置,默认工作线程20个
for (int i = 0; i < workerCount; i++) {
// 回调线程创建实现
WorkerRunnable<ID, T> runnable = workerRunnableFactory.create(i);
// 创建线程
Thread workerThread = new Thread(threadGroup, runnable, runnable.getWorkerName());
workerThreads.add(workerThread);
workerThread.setDaemon(true);
// 启动线程
workerThread.start();
}
}
// 创建批量任务执行器工具类
static <ID, T> TaskExecutors<ID, T> batchExecutors(final String name,
int workerCount,
final TaskProcessor<T> processor,
final AcceptorExecutor<ID, T> acceptorExecutor) {
final AtomicBoolean isShutdown = new AtomicBoolean();
// 任务执行指标
final TaskExecutorMetrics metrics = new TaskExecutorMetrics(name);
// 创建TaskExecutors对象,并创建线程工厂WorkerRunnableFactory。创建指定的工作线程BatchWorkerRunnable
return new TaskExecutors<>(new WorkerRunnableFactory<ID, T>() {
// 线程创建的实现
@Override
public WorkerRunnable<ID, T> create(int idx) {
// 线程名:TaskBatchingWorker-target_主机名-索引下标
return new BatchWorkerRunnable<>("TaskBatchingWorker-" +name + '-' + idx, isShutdown, metrics, processor, acceptorExecutor);
}
}, workerCount, isShutdown);
}
👍👍小结:
① SpringCloud 会接管 Eureka Server端得一系列组件创建过程,包括其中重要得组件:EurekaServerContext、EurekaServerBootstrap、PeerEurekaNodes、ResponseCache、EurekaServerConfig、PeerAwareInstanceRegistry、ApplicationInfoManager等。
② 扩展了PeerEurekaNodes得实现 RefreshablePeerEurekaNodes(可刷新得Eureka 节点,即当环境参数变化时(监听到 EnvironmentChangeEvent事件)就会触发Eureka Server的更新服务列表逻辑)。然后默认的实现DefaultEurekaServerContext 会启动一个每隔10分钟得定时任务,也会去更新服务列表
③ 通过EurekaServerBootstrap类来管理Eureka Server节点 环境和上下文的生命周期
④ 通过PeerAwareInstanceRegistry来实现Eureka Server节点的相关操作(注册、取消、续约、节点复制、更新等)
⑤ 通过EurekaServerConfig来管理Eureka Server端节点的属性配置
⑥ 通过PeerEurekaNodes类来管理节点的生命周期(启动、关闭、更新、新增)
⑦ 通过ResponseCache来缓存服务实例(readWriteCacheMap、readOnlyCacheMap),默认30秒会定时从readWriteCacheMap中将数据同步到readOnlyCacheMap中
⑧ 通过PeerEurekaNode服务端节点,执行peer节点得复制、批量发送等操作。封装了一系列得服务端交互接口:包括服务注册、心跳检测、服务下线、状态更新。其中,批量任务处理器得创建又会引入两个线程:BatchWorkerRunnable和AcceptorRunner,这两个线程通过生产者消费者模型(batchWorkQueue)实现任务处理与处理结果获取。还实现了一些异常处理机制(失败重试等):reprocessQueue、acceptorQueue、pendingTasks。(这块得批量任务处理方式很值得我们学习,另外Kafka得批量消息处理也很值得我们借鉴与吸收)
- 到此,Eureka服务端已经大致分析完毕,下篇将会针对于Eureka客户端进行剖析。
- ☛ 文章要是勘误或者知识点说的不正确,欢迎评论,毕竟这也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!
网友评论