聊聊微服务熔断降级Hystrix
在现在的微服务使用的过程中,经常会遇到依赖的服务不可用,那么如果依赖的服务不可用的话,会导致把自己的服务也会拖死,那么就产生了熔断,熔断顾名思义就是当服务处于不可用的时候采取半开关的状态,达到一定数量后就熔断器就打开。这就相当于家里边的保险丝,如果电压过高的话,保险丝就会断掉,起到保护电器的作用。
目前支持熔断,降级的就是Hystrix,当然还有resilience4j还有Sentinel。今天咱们以Hystrix为主吧。其他的大家可以自行研究。
Hystrix主要实现三个功能,接下来咱们继续展开。
1. 资源隔离
2. 熔断
3. 降级
资源隔离分为两种,一种是线程池隔离,一种是信号量semaphore隔离。线程池以请求的线程和执行的线程分为不同的线程执行,信号量是请求和执行采用相同的线程。
当然,涉及到线程池的话,那么就支持异步,支持异步Future的话也就支持get的时候支持超时获取。信号量这些功能不支持,但是信号量是支持熔断,限流。他们的区别如下:
线程切换 | 异步 | 超时 | 熔断 | 限流 | 资源开销 | |
线程池 | 是 | 是 | 是 | 是 | 是 | 大 |
信号量 | 否 | 否 | 否 | 是 | 是 | 小 |
HystrixCommand的命令执行大致如下图:
依赖的pom如下:
<!-- 依赖版本 --> <hystrix.version>1.3.16</hystrix.version> <hystrix-metrics-event-stream.version>1.1.2</hystrix-metrics-event-stream.version> <dependency> <groupId>com.netflix.hystrix</groupId> <artifactId>hystrix-core</artifactId> <version>${hystrix.version}</version> </dependency> <dependency> <groupId>com.netflix.hystrix</groupId> <artifactId>hystrix-metrics-event-stream</artifactId> <version>${hystrix-metrics-event-stream.version}</version> </dependency>
支持同步,异步,观察事件拦截,以及订阅方式,下面咱们直接看代码实现吧。大家一看就明白了:
import com.netflix.hystrix.HystrixCommand; import com.netflix.hystrix.HystrixCommandGroupKey; import rx.Observable; import rx.Subscriber; import rx.functions.Action1; import java.util.concurrent.ExecutionException; import java.util.concurrent.Future; import java.util.concurrent.TimeUnit; import java.util.concurrent.TimeoutException; /** * @author huangqingshi * @Date 2019-03-17 */ public class HelloWorldCommand extends HystrixCommand<String> { private final String name; public HelloWorldCommand(String name) { //指定命令组名 super(HystrixCommandGroupKey.Factory.asKey("myGroup")); this.name = name; } @Override protected String run() throws Exception { //逻辑封装在run里边 return "Hello:" + name + " thread:" + Thread.currentThread().getName(); } public static void main(String[] args) throws InterruptedException, ExecutionException, TimeoutException { //每个Command只能调用一次,不能重复调用。重复调用会报异常。 HelloWorldCommand helloWorldCommand = new HelloWorldCommand("Synchronous-hystrix"); //execute同步调用 等同于:helloWorldCommand.queue().get(); String result = helloWorldCommand.execute(); System.out.println("result:" + result); helloWorldCommand = new HelloWorldCommand("Asynchronous-hystrix"); //异步调用 Future<String> future = helloWorldCommand.queue(); //get可以指定获取的时间100毫秒,默认为1秒 result = future.get(100, TimeUnit.MILLISECONDS); System.out.println("result:" + result); System.out.println("main thread:" + Thread.currentThread().getName()); testObserve(); } public static void testObserve() { //注册观察者事件拦截 Observable<String> observable = new HelloWorldCommand("observe").observe(); //注册回调事件 observable.subscribe(new Action1<String>() { @Override public void call(String result) { //result就是调用HelloWorldCommand的结果 System.out.println("callback:" + result); } }); //注册完成版的事件 observable.subscribe(new Subscriber<String>() { @Override public void onCompleted() { System.out.println("onCompleted调用:onNext : onError之后调用"); } @Override public void onError(Throwable throwable) { //异常产生了之后会调用 System.out.println("onError:" + throwable.getMessage()); } @Override public void onNext(String s) { //获取结果后回调 System.out.println("onNext:" + s); } }); } }
执行结果如下:
result:Hello:Synchronous-hystrix thread:hystrix-myGroup-1
result:Hello:Asynchronous-hystrix thread:hystrix-myGroup-2
main thread:main
callback:Hello:observe thread:hystrix-myGroup-3
onNext:Hello:observe thread:hystrix-myGroup-3
onCompleted调用:onNext : onError之后调用
接下来是线程池隔离的例子:
import com.netflix.hystrix.*; /** * @author huangqingshi * @Date 2019-03-17 */ public class ThreadPoolCommand extends HystrixCommand<String> { private String name; public ThreadPoolCommand(String name) { super(Setter.withGroupKey(HystrixCommandGroupKey.Factory.asKey("threadPoolGroup")) .andCommandKey(HystrixCommandKey.Factory.asKey("threadPoolCommand")) .andCommandPropertiesDefaults(HystrixCommandProperties.Setter() .withCircuitBreakerRequestVolumeThreshold(10) //至少10个请求,熔断器才进行错误计算 默认值20 .withCircuitBreakerSleepWindowInMilliseconds(5000) //熔断终端5秒后会进入半打开状态 .withCircuitBreakerErrorThresholdPercentage(50) //错误率达到50开启熔断保护 .withExecutionIsolationStrategy(HystrixCommandProperties.ExecutionIsolationStrategy.THREAD) //10个核心线程 ).andThreadPoolPropertiesDefaults(HystrixThreadPoolProperties.Setter().withCoreSize(10)) ); this.name = name; } @Override protected String run() throws Exception { return "threadPoolCommand name:" + name; } public static void main(String[] args) { ThreadPoolCommand threadPoolCommand = new ThreadPoolCommand("threadPool"); String result = threadPoolCommand.execute(); System.out.println("result:" + result); } } 执行结果: result:threadPoolCommand name:threadPool
信号量隔离例子:
/** * @author huangqingshi * @Date 2019-03-17 */ public class SemaphoreCommand extends HystrixCommand<String> { private String name; public SemaphoreCommand(String name) { super(Setter.withGroupKey(HystrixCommandGroupKey.Factory.asKey("semaphoreGroup")) .andCommandKey(HystrixCommandKey.Factory.asKey("semaphoreCommand")) .andCommandPropertiesDefaults(HystrixCommandProperties.Setter() //至少10个请求,熔断器才会进行错误率的计算 默认值20 .withCircuitBreakerRequestVolumeThreshold(10) //熔断器中断请求5秒后会自动进入半打开状态,放部分流量进行重试 默认值5000ms .withCircuitBreakerSleepWindowInMilliseconds(5000) //错误率达到50开启熔断保护 .withCircuitBreakerErrorThresholdPercentage(50) //设置隔离策略 .withExecutionIsolationStrategy(HystrixCommandProperties.ExecutionIsolationStrategy.SEMAPHORE) //最大并发量10 .withExecutionIsolationSemaphoreMaxConcurrentRequests(10) ) ); this.name = name; } @Override protected String run() throws Exception { return "semaphore success name:" + name; } @Override protected String getFallback() { return "semaphore fallback name:" + name; } public static void main(String[] args) { SemaphoreCommand semaphoreCommand = new SemaphoreCommand("semaphoreCommand"); String result = semaphoreCommand.execute(); System.out.println(result); } } 执行结果: semaphore success name:semaphoreCommand
在执行的过程中,如果出现调用服务的时候出现错误的时候会先进行熔断,就是如果流量达到设置的量的时候进行统计,比如10个请求,然后如果出现错误率超过配置的错误率就会进行将熔断进行打开,打开之后会进行调用降级方法fallback。过了一段时间后,可以放行部分流量,如果流量正常了,则会将熔断器开关关闭。下图是来自官方文档截图,里边维护者一个bucket,每秒一个bucket,里边记录着成功,失败,超时,拒绝。这个周期是通过withCircuitBreakerSleepWindowInMilliseconds配置的。
接下来咱们看一下降级,也就是熔断器打开的时候,会走fallback方法,继续看例子。
import com.netflix.hystrix.*; /** * @author huangqingshi * @Date 2019-03-17 */ public class ThreadPoolCommand extends HystrixCommand<String> { private String name; public ThreadPoolCommand(String name) { super(Setter.withGroupKey(HystrixCommandGroupKey.Factory.asKey("threadPoolGroup")) .andCommandKey(HystrixCommandKey.Factory.asKey("threadPoolCommand")) .andCommandPropertiesDefaults(HystrixCommandProperties.Setter() .withCircuitBreakerRequestVolumeThreshold(10) //至少10个请求,熔断器才进行错误计算 默认值20 .withCircuitBreakerSleepWindowInMilliseconds(5000) //熔断终端5秒后会进入半打开状态 .withCircuitBreakerErrorThresholdPercentage(50) //错误率达到50开启熔断保护 .withExecutionIsolationStrategy(HystrixCommandProperties.ExecutionIsolationStrategy.THREAD) //10个核心线程 ).andThreadPoolPropertiesDefaults(HystrixThreadPoolProperties.Setter().withCoreSize(10)) ); this.name = name; } @Override protected String run() throws Exception { return "threadPoolCommand name:" + name; } public static void main(String[] args) { ThreadPoolCommand threadPoolCommand = new ThreadPoolCommand("threadPool"); String result = threadPoolCommand.execute(); System.out.println("result:" + result); } } 执行结果: result:executed fallback 并且抛出超时异常。因为程序故意设计超时的。
当然,Hystrixcommand还支持primary或secondary的方式,可以先看看流程图:
是否执行primary是通过参数primarySecondary.userPrimary为true时执行。false的时候执行secondary方式。
/** * @author huangqingshi * @Date 2019-03-18 */ public class PrimarySecondaryFacade extends HystrixCommand<String> { private final static DynamicBooleanProperty usePrimary = DynamicPropertyFactory.getInstance(). getBooleanProperty("primarySecondary.usePrimary", true); private int id; public PrimarySecondaryFacade(int id) { super(Setter.withGroupKey(HystrixCommandGroupKey.Factory.asKey("facadeGroup")) .andCommandKey(HystrixCommandKey.Factory.asKey("primarySecondCommand")) //此处采用信号量,primary、secondary采用线程池 .andCommandPropertiesDefaults(HystrixCommandProperties.Setter().withExecutionIsolationStrategy( HystrixCommandProperties.ExecutionIsolationStrategy.SEMAPHORE) ) ); this.id = id; } @Override protected String run() throws Exception { if(usePrimary.get()) { return new PrimaryCommand(id).execute(); } else { return new SecondaryCommand(id).execute(); } } @Override protected String getFallback() { return "static-fallback-" + id; } @Override protected String getCacheKey() { return String.valueOf(id); } private static class PrimaryCommand extends HystrixCommand<String> { private final int id; private PrimaryCommand(int id) { super(Setter .withGroupKey(HystrixCommandGroupKey.Factory.asKey("facadeGroup")) .andCommandKey(HystrixCommandKey.Factory.asKey("PrimaryCommand")) .andThreadPoolKey(HystrixThreadPoolKey.Factory.asKey("PrimaryCommand")) .andCommandPropertiesDefaults(HystrixCommandProperties.Setter(). withExecutionIsolationThreadTimeoutInMilliseconds(600))); this.id = id; } @Override protected String run() { return "responseFromPrimary-" + id; } } private static class SecondaryCommand extends HystrixCommand<String> { private final int id; private SecondaryCommand(int id) { super(Setter .withGroupKey(HystrixCommandGroupKey.Factory.asKey("facadeGroup")) .andCommandKey(HystrixCommandKey.Factory.asKey("SecondaryCommand")) .andThreadPoolKey(HystrixThreadPoolKey.Factory.asKey("SecondaryCommand")) .andCommandPropertiesDefaults(HystrixCommandProperties.Setter(). withExecutionIsolationThreadTimeoutInMilliseconds(600))); this.id = id; } @Override protected String run() { return "responseFromSecondary-" + id; } } public static class UnitTest { @Test public void testPrimary() { HystrixRequestContext context = HystrixRequestContext.initializeContext(); try { ConfigurationManager.getConfigInstance().setProperty("primarySecondary.usePrimary", true); assertEquals("responseFromPrimary-100", new PrimarySecondaryFacade(100).execute()); } finally { context.shutdown(); ConfigurationManager.getConfigInstance().clear(); } } @Test public void testSecondary() { HystrixRequestContext context = HystrixRequestContext.initializeContext(); try { ConfigurationManager.getConfigInstance().setProperty("primarySecondary.usePrimary", false); assertEquals("responseFromSecondary-100", new PrimarySecondaryFacade(100).execute()); } finally { context.shutdown(); ConfigurationManager.getConfigInstance().clear(); } } } }
好了,这个基本上就是Hystrix的基本功能,但是有个问题就是Hystrix已经不维护了,但是目前的稳定版大家也都在使用,所以列出来了。当然也推荐大家使用Sentinel,功能比较强大,就是自适应限流功能等,功能也非常强大,后续研究之后再出相关文章吧。这个文章就当大家的一个敲门砖吧,有问题请及时告知,谢谢。