Hystrix (容错，回退，降级，缓存)

Hystrix熔断机制就像家里的保险丝一样，若同时使用高功率的电器，就会烧坏电路，这时候保险丝自动断开就有效的保护了电路。而我们程序中也同样是这样。例如若此时数据库压力太大速度很慢，此时还有不断的请求访问后台，就会造成数据库崩溃。这时候hystrix容错机制，可以为客户端请求设置超时链接，添加回退的逻辑，减少集群压力。

1.1 导依赖

    <parent>
        <groupId>org.springframework.boot</groupId>
        <artifactId>spring-boot-starter-parent</artifactId>
        <version>1.5.8.RELEASE</version>
        <relativePath/>
    </parent>
    <!-- springCloud -->
    <dependencyManagement>
        <dependencies>
            <dependency>
                <groupId>org.springframework.cloud</groupId>
                <artifactId>spring-cloud-dependencies</artifactId>
                <version>Dalston.SR3</version>
                <type>pom</type>
                <scope>import</scope>
            </dependency>
        </dependencies>
    </dependencyManagement>

    <dependencies>
        <dependency>
            <groupId>org.springframework.boot</groupId>
            <artifactId>spring-boot-starter-web</artifactId>
        </dependency>
        <!-- eureka客户端依赖 -->
        <dependency>
            <groupId>org.springframework.cloud</groupId>
            <artifactId>spring-cloud-starter-eureka</artifactId>
        </dependency>
        <!-- hystrix -->
        <dependency>
            <groupId>org.springframework.cloud</groupId>
            <artifactId>spring-cloud-starter-hystrix</artifactId>
        </dependency>

1.2 配置application.yml

# 指定端口
server:
  port: 8085
  context-path: /demo
# 服务名称
spring:
  application:
    name: hystrix

# eureka服务器地址
eureka:
  client:
    serviceUrl:
      defaultZone: http://localhost:8761/eureka/

1.3 配置Hystrix过滤器

import java.io.IOException;
import javax.servlet.Filter;
import javax.servlet.FilterChain;
import javax.servlet.FilterConfig;
import javax.servlet.ServletException;
import javax.servlet.ServletRequest;
import javax.servlet.ServletResponse;
import javax.servlet.annotation.WebFilter;
import com.netflix.hystrix.strategy.concurrency.HystrixRequestContext;
/**
 * urlPatterns:拦截所有路径(拦截规则)
 * filterName:过滤器名称
 */
@WebFilter(urlPatterns = "/*", filterName = "hystrixFilter")
public class HystrixFilterConf implements Filter{
    public void destroy() { }

    public void doFilter(ServletRequest request, ServletResponse response, FilterChain chain)
            throws IOException, ServletException {
        // 初始化Hystrix上下文
        HystrixRequestContext ctx = HystrixRequestContext.initializeContext();
        try {
            chain.doFilter(request, response);
        } catch (Exception e) {

        } finally {
            ctx.shutdown();
        }
    }
    public void init(FilterConfig arg0) throws ServletException { }
}

1.4 主函数入口

import org.springframework.boot.autoconfigure.SpringBootApplication;
import org.springframework.boot.builder.SpringApplicationBuilder;
import org.springframework.boot.web.servlet.ServletComponentScan;
import org.springframework.cloud.client.circuitbreaker.EnableCircuitBreaker;
import org.springframework.cloud.netflix.eureka.EnableEurekaClient;
@SpringBootApplication
@EnableEurekaClient  //开启Eureka
@EnableCircuitBreaker//开启断路器
@ServletComponentScan//扫描servlet过滤器监听器
public class ServerMain {

    public static void main(String[] args) {
        new SpringApplicationBuilder(ServerMain.class).web(true).run(args);
    }
}

二： hystrix 的 熔断，降级 机制。

2.1 回退机制案例

import com.netflix.hystrix.contrib.javanica.annotation.HystrixCommand;
import com.netflix.hystrix.contrib.javanica.annotation.HystrixProperty;
import org.springframework.stereotype.Service;
import org.springframework.web.bind.annotation.RequestMapping;
/**
 * Hystrix回退机制(熔断,降级)
 */
@Service
public class RollbackService {

    /**
     * fallbackMethod: 指定回退方法
     *
     * coreSize: 线程池最大数量
     * maxQueueSize: 线程池最大队列,默认-1通过SynchronousQueue来实现; 否则使用LinkedBlockingQueue实现
     * queueSizeRejectionThreshold: 当maxQueueSize是LinkedBlockingQueue时,即使没有达到最大列队也会据绝请求。
     *
     * timeoutInMilliseconds: 超时时间
     * requestVolumeThreshold: 单位时间内超过这个多个请求失败才执行熔断
     */
    @RequestMapping(value = "/testRollback")
    @HystrixCommand(fallbackMethod = "myRollback",
        threadPoolProperties = {
            @HystrixProperty(name = "coreSize", value = "30"),
            @HystrixProperty(name = "maxQueueSize", value = "-1"),
            @HystrixProperty(name = "queueSizeRejectionThreshold", value = "-1")
        },
        commandProperties = {
            @HystrixProperty(name = "execution.isolation.thread.timeoutInMilliseconds", value = "3000"),
            @HystrixProperty(name = "circuitBreaker.requestVolumeThreshold", value = "1")
        })
    public String testRollback() {
        try {// 模拟请求阻塞
            Thread.sleep(4000);
        } catch (Exception e) {
        }
        return "恭喜你访问成功!";
    }
    /** 回退方法  */
    public String myRollback() {
        return "服务器压力过大,明年再来吧!";
    }
}

2.2 编写接口， 调用测试。

import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.web.bind.annotation.RequestMapping;
import org.springframework.web.bind.annotation.RestController;
@RestController
public class RollbackController {

    @Autowired
    private RollbackService rollbackService;

    @RequestMapping("/testRollback")
    public String testRollback() {
        return rollbackService.testRollback();
    }
}

三： hystrix 缓存机制：hystrix的缓存有点像mybatis默认开启的一级缓存：在session关闭之前(一次会话期间)，使用同样的参数调用同一个方法，实际只查询一次。

3.1 缓存逻辑

import org.springframework.stereotype.Service;
import com.netflix.hystrix.contrib.javanica.annotation.HystrixCommand;
import com.netflix.hystrix.contrib.javanica.cache.annotation.CacheRemove;
import com.netflix.hystrix.contrib.javanica.cache.annotation.CacheResult;
@Service
public class CacheService {

    @CacheResult
    @HystrixCommand
    public void cacheMember(Integer id) {
        System.out.println("调用 cacheMember 方法");
    }

    /**
     * commandKey:缓存的key
     *       获取和删除必须用同一个key,并必须是同一次请求。
     */
    @CacheResult
    @HystrixCommand(commandKey = "myCacheKey")
    public void getCache(Integer id) {
        System.out.println("执行查询方法");
    }
    @CacheRemove(commandKey = "myCacheKey")
    @HystrixCommand
    public void removeCache(Integer id) {
        System.out.println("删除缓存方法");
    }
}

3.2 缓存接口开发

import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.http.MediaType;
import org.springframework.web.bind.annotation.RequestMapping;
import org.springframework.web.bind.annotation.RequestMethod;
import org.springframework.web.bind.annotation.RestController;
@RestController
public class CacheController {

    @Autowired
    private CacheService cacheService;

    @RequestMapping(value = "/cache", method = RequestMethod.GET,
            produces = MediaType.APPLICATION_JSON_VALUE)
    public String cache() {
        for(int i = 0; i < 3; i++) {
            /** 在同一次请求里面调用了3次同一方法，会发现，控制台只
             *  输出了一次，说明后2次走的缓存没调方法的逻辑    */
            cacheService.cacheMember(1);
        }
        System.out.println("测试完毕");
        return "";
    }
    @RequestMapping(value = "/rc", method = RequestMethod.GET,
            produces = MediaType.APPLICATION_JSON_VALUE)
    public String testRemoveCache() {
        cacheService.getCache(1);
        cacheService.getCache(1);

        cacheService.removeCache(1);
        System.out.println("#########  分隔线   ###########");
        cacheService.getCache(1);
        System.out.println("测试完毕");
        return "";
    }
}