Spring Cloud之Hystrix
在微服务架构中,存在那么多的服务单元,若一个单元出现故障(由于网络原因或者自身原因),就很容易因依赖关系而引发故障的蔓延,最终导致整个系统的瘫痪,这样的架构相较传统架构更加不稳定。为了解决这样的问题,产生了断路器等一系列的服务保护机制。(A服务调用B服务,B服务由于自身处理逻辑等原因造成响应缓慢,会导致A服务线程被挂起,以等待B服务执行,在高并发情况下,这些挂起的线程会导致后面调用A服务的请求被阻塞,最终导致A服务也不可用)。
加入断路器后,当服务不可用时,通过断路器的故障监控,会直接执行回调函数,直接返回一串字符串,而不是等待响应超时,这样就不会使得线程调用故障服务被长时间占用不释放,从而避免了故障在分布式系统中的蔓延。
本节内容在上节内容基础上,阅读此节之前,先看上节Spring Cloud之Eureka、Ribbon
一、无断路器示例
启动上节的eureka-server、service-hello(8081/8082)、ribbon-consumer工程
在未加入断路器之前,关闭8081实例,发送GET请求到http://localhost:9000/ribbon-consumer,轮询8081/8082,当轮询到8081后(因为8081实例被关)会得到下面输出:
二、加入断路器示例
在ribbon-consumer工程的pom.xml引入下面依赖
<dependency> <groupId>org.springframework.cloud</groupId> <artifactId>spring-cloud-starter-hystrix</artifactId> </dependency>
在ribbon-consumer工程的 主类ConsumerApplication中使用@EnableCircuitBreaker注解开启断路器功能
package com.stonegeek; import org.springframework.boot.SpringApplication; import org.springframework.boot.autoconfigure.SpringBootApplication; import org.springframework.cloud.client.circuitbreaker.EnableCircuitBreaker; import org.springframework.cloud.client.discovery.EnableDiscoveryClient; import org.springframework.cloud.client.loadbalancer.LoadBalanced; import org.springframework.context.annotation.Bean; import org.springframework.web.client.RestTemplate; /** * Created by StoneGeek on 2018/5/28. * 博客地址:http://www.cnblogs.com/sxkgeek */ @EnableCircuitBreaker @EnableDiscoveryClient @SpringBootApplication public class RibbonConsumerApplication { @Bean @LoadBalanced RestTemplate restTemplate(){ return new RestTemplate(); } public static void main(String[] args) { SpringApplication.run(RibbonConsumerApplication.class, args); } }
改造服务消费方式,新增HelloService类,注入RestTemplate实例,然后将在ConsumerController中对RestTemplate的使用迁移到helloService函数中,最后,在helloService函数上增加@HystrixCommand注解来指定回调方法:
package com.stonegeek.service; import com.netflix.hystrix.contrib.javanica.annotation.HystrixCommand; import org.springframework.beans.factory.annotation.Autowired; import org.springframework.stereotype.Service; import org.springframework.web.client.RestTemplate; import java.util.logging.Logger; /** * Created by StoneGeek on 2018/5/29. * 博客地址:http://www.cnblogs.com/sxkgeek */ @Service public class HelloService { private final Logger logger =Logger.getLogger(String.valueOf(getClass())); @Autowired RestTemplate restTemplate; @HystrixCommand(fallbackMethod = "helloFallback") public String helloService(){ //加logger更清晰的看出执行时间 long start =System.currentTimeMillis(); String result=restTemplate.getForEntity("http://SERVICE-HELLO/hello",String.class).getBody(); long end=System.currentTimeMillis(); logger.info("Spend time:"+(end-start)); return result; } public String helloFallback(){ return "error"; } }
修改RibbonConsumerApplication.class
package com.stonegeek.controller; import com.stonegeek.service.HelloService; import org.springframework.beans.factory.annotation.Autowired; import org.springframework.web.bind.annotation.RequestMapping; import org.springframework.web.bind.annotation.RequestMethod; import org.springframework.web.bind.annotation.RestController; import org.springframework.web.client.RestTemplate; /** * Created by StoneGeek on 2018/5/28. * 博客地址:http://www.cnblogs.com/sxkgeek */ @RestController public class ConsumerController { /** * @Author: StoneGeek * @Date: 2018/5/29 * @Description:之前不加断路器代码 */ // @Autowired // RestTemplate restTemplate; // // @RequestMapping(value = "/ribbon-consumer",method = RequestMethod.GET) // public String helloConsumer(){ // return restTemplate.getForEntity("http://SERVICE-HELLO/hello",String.class).getBody(); // } /** * @Author: StoneGeek * @Date: 2018/5/29 * @Description: 加了断路器代码 */ @Autowired HelloService helloService; @RequestMapping(value = "/ribbon-consumer",method = RequestMethod.GET) public String helloConsumer(){ return helloService.helloService(); } }
此时重新来验证一下断路器实现的服务回调逻辑,此时断开8081实例,当服务消费者轮询到8081时,不再是之前的错误内容,Hystrix服务回调生效
三、模拟服务阻塞来验证断路器回调
Hystrix默认超时时间是2000毫秒
我们对service-hello的/hello接口做一些修改(重点是Thread.sleep()函数的使用):
package com.stonegeek.controller; import org.springframework.beans.factory.annotation.Autowired; import org.springframework.cloud.client.ServiceInstance; import org.springframework.cloud.client.discovery.DiscoveryClient; import org.springframework.web.bind.annotation.RequestMapping; import org.springframework.web.bind.annotation.RequestMethod; import org.springframework.web.bind.annotation.RestController; import java.util.ArrayList; import java.util.List; import java.util.Random; import java.util.logging.Logger; /** * Created by StoneGeek on 2018/5/27. * 博客地址:http://www.cnblogs.com/sxkgeek */ @RestController public class HelloController { private final Logger logger =Logger.getLogger(String.valueOf(getClass())); @Autowired private DiscoveryClient client; /** * @Author: StoneGeek * @Date: 2018/5/29 * @Description:不加线服务塞的代码 */ // @RequestMapping(value = "hello",method = RequestMethod.GET) // public String index(){ // ServiceInstance instance=client.getLocalServiceInstance(); // logger.info("/hello, host:"+instance.getHost()+", service_id:"+instance.getServiceId()+"service_port:"+instance.getPort()); // return "hello world "; // } /** * @Author: StoneGeek * @Date: 2018/5/29 * @Description: 加了Thread.sleep(3000)的服务阻塞代码,由于Hystrix默认超时时间为2000毫秒, * 所以这里采用了0至3000的随机数以让处理过程有一定概率发生超时来触发断路器 */ @RequestMapping(value = "/hello",method = RequestMethod.GET) public String index() throws InterruptedException { ServiceInstance instance=client.getLocalServiceInstance(); int sleepTime=new Random().nextInt(3000); logger.info("sleepTime:"+sleepTime); Thread.sleep(sleepTime); logger.info("sleepTime:"+sleepTime+"/hello, host:"+instance.getHost()+", service_id:"+instance.getServiceId()+"service_port:"+instance.getPort()); return "hello world "; } }
重新启动service-hello和ribbon-consumer模块,连续访问http://localhost:9000/ribbon-consumer几次,当RIBBON-CONSUMER的控制台输出的Spend time大于2000的时候,网页就会返回error,即服务消费者因调用的服务超时从而触发熔断请求,并调用回调逻辑返回结果
此时Spring Could Hystrix的断路器就配置完成了!!!