Java基础之《微服务（10）—消息驱动》
一、什么是bus
1、什么是spring cloud bus
spring cloud bus集成了市面上常用的消息代理（rabbit mq、kafka等2种），连接微服务系统中的所有节点，当有数据变更时，可以通过消息代理广播通知微服务及时变更数据。例如微服务的配置更新。
2、bus解决了什么问题
解决了微服务数据变更，及时同步的问题。
二、消息发送模块
1、复制或者新建一个maven模块mycloud-stream-sender
2、pom文件
添加rabbit依赖包spring-cloud-starter-stream-rabbit
4.0.0
org.springframework.boot
spring-boot-starter-parent
2.1.8.RELEASE
com.example
mycloud-stream-sender
0.0.1-SNAPSHOT
mycloud-stream-sender
Demo project for Spring Boot
1.8
Greenwich.SR5
org.springframework.boot
spring-boot-starter
org.springframework.boot
spring-boot-starter-test
test
org.springframework.cloud
spring-cloud-starter-netflix-eureka-client
org.springframework.boot
spring-boot-starter-web
org.springframework.boot
spring-boot-starter-actuator
org.springframework.cloud
spring-cloud-starter-stream-rabbit
org.springframework.cloud
spring-cloud-dependencies
${spring-cloud.version}
pom
import
org.springframework.boot
spring-boot-maven-plugin
3、发送接口
ISendService.java
package com.example.mycloud.resource;
import org.springframework.cloud.stream.annotation.Output;
import org.springframework.messaging.SubscribableChannel;
/**
* 发送的接口
* @author user
*
*/
public interface ISendService {
@Output("stream-exchange")
public SubscribableChannel send();
}4、测试类
TestSender.java
package com.example.mycloud.resource;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.messaging.Message;
import org.springframework.messaging.support.MessageBuilder;
import org.springframework.web.bind.annotation.RequestMapping;
import org.springframework.web.bind.annotation.RestController;
@RestController
public class TestSender {
@Autowired
private ISendService sendService;
@RequestMapping("/send")
public void send() {
String msg = "abc..........";
Message message = MessageBuilder.withPayload(msg.getBytes()).build();
sendService.send().send(message);
}
}5、启动类
StreamSenderApplication.java
package com.example.mycloud.run;
import org.springframework.boot.SpringApplication;
import org.springframework.boot.autoconfigure.SpringBootApplication;
import org.springframework.cloud.client.discovery.EnableDiscoveryClient;
import org.springframework.cloud.stream.annotation.EnableBinding;
import org.springframework.context.annotation.ComponentScan;
import com.example.mycloud.resource.ISendService;
@EnableBinding({ISendService.class}) //把发送接口绑定进来
@EnableDiscoveryClient
@SpringBootApplication
@ComponentScan("com.example.mycloud")
public class StreamSenderApplication {
public static void main(String[] args) {
SpringApplication.run(StreamSenderApplication.class, args);
}
}6、配置文件
spring:
application:
name: stream-sender
rabbitmq:
host: 127.0.0.1
port: 5672
username: admin
password: admin
virtual-host: my_vhost
server:
port: 8017
eureka:
server:
port: 8010
instance:
hostname: localhost
client:
registerWithEureka: true
fetchRegistry: true
serviceUrl:
defaultZone: http://admin:123456@${eureka.instance.hostname}:${eureka.server.port}/eureka/
#暴露actuator的所有端口
management:
endpoints:
web:
exposure:
include: "*"
endpoint:
health:
show-details: ALWAYS
shutdown:
#启用shutdown
enabled: true
#禁用密码验证
sensitive: false
三、消息接收模块
1、复制或者新建一个maven模块mycloud-stream-receiver
2、pom文件
4.0.0
org.springframework.boot
spring-boot-starter-parent
2.1.8.RELEASE
com.example
mycloud-stream-receiver
0.0.1-SNAPSHOT
mycloud-stream-receiver
Demo project for Spring Boot
1.8
Greenwich.SR5
org.springframework.boot
spring-boot-starter
org.springframework.boot
spring-boot-starter-test
test
org.springframework.cloud
spring-cloud-starter-netflix-eureka-client
org.springframework.boot
spring-boot-starter-web
org.springframework.boot
spring-boot-starter-actuator
org.springframework.cloud
spring-cloud-starter-stream-rabbit
org.springframework.cloud
spring-cloud-dependencies
${spring-cloud.version}
pom
import
org.springframework.boot
spring-boot-maven-plugin
3、接收接口
IReceiveService.java
package com.example.mycloud.resource;
import org.springframework.cloud.stream.annotation.Input;
import org.springframework.messaging.SubscribableChannel;
/**
* 接收的接口
* @author user
*
*/
public interface IReceiveService {
@Input("stream-exchange")
public SubscribableChannel receive();
}4、监听类
ReceiveService.java
package com.example.mycloud.service;
import org.springframework.cloud.stream.annotation.EnableBinding;
import org.springframework.cloud.stream.annotation.StreamListener;
import org.springframework.stereotype.Service;
import com.example.mycloud.resource.IReceiveService;
@Service
@EnableBinding(IReceiveService.class) //绑定
public class ReceiveService {
@StreamListener("stream-exchange")
public void onReceive(byte[] msg) {
System.out.println("receive: " + new String(msg));
}
}5、启动类
StreamReceiverApplication.java
package com.example.mycloud.run;
import org.springframework.boot.SpringApplication;
import org.springframework.boot.autoconfigure.SpringBootApplication;
import org.springframework.cloud.client.discovery.EnableDiscoveryClient;
import org.springframework.context.annotation.ComponentScan;
@EnableDiscoveryClient
@SpringBootApplication
@ComponentScan("com.example.mycloud")
public class StreamReceiverApplication {
public static void main(String[] args) {
SpringApplication.run(StreamReceiverApplication.class, args);
}
}6、配置文件
spring:
application:
name: stream-receiver
rabbitmq:
host: 127.0.0.1
port: 5672
username: admin
password: admin
virtual-host: my_vhost
server:
port: 8018
eureka:
server:
port: 8010
instance:
hostname: localhost
client:
registerWithEureka: true
fetchRegistry: true
serviceUrl:
defaultZone: http://admin:123456@${eureka.instance.hostname}:${eureka.server.port}/eureka/
#暴露actuator的所有端口
management:
endpoints:
web:
exposure:
include: "*"
endpoint:
health:
show-details: ALWAYS
shutdown:
#启用shutdown
enabled: true
#禁用密码验证
sensitive: false四、stream解决了什么问题
1、应用图
Application Core：应用服务
2、stream解决了开发人员无感知使用消息中间件的问题
因为stream对消息中间件的进一步封装，可以做到代码层面对消息中间件的无感知，甚至于动态的切换中间件（例如从rabbitmq切换为kafka）。使得微服务开发的高度解耦，服务可以关注更多自己的业务流程。
3、middleware
消息中间件，目前只支持rabbitmq和kafka。
4、binder
binder是应用与消息中间件之间的封装。目前实现了kafka和rabbitmq的binder。通过binder，可以很方便的连接中间件，可以动态改变消息类型。
5、@Input
注解标识输入通道，通过该输入通道接收到的消息进入应用程序。
6、@Output
注解标识输出通道，发布的消息将通过该通道离开应用程序。
7、@StreamListener
监听队列，用于消费者的队列的消息接收。
8、@EnableBinding
指信道channel和exchange绑定在一起。
五、消息的分组
1、之前的服务有什么问题
（1）队列是临时队列
（2）如果起多个receiver就会有多个队列queue绑定到exchange上，多个服务会同时接收到消息，消费多次
2、指定信道绑定交换器，交换器绑定队列
stream-sender模块
ISendService.java添加：
package com.example.mycloud.resource;
import org.springframework.cloud.stream.annotation.Output;
import org.springframework.messaging.SubscribableChannel;
/**
* 发送的接口
* @author user
*
*/
public interface ISendService {
@Output("stream-exchange")
public SubscribableChannel send();
//消息分组
String OUTPUT = "groupOutput";
@Output(ISendService.OUTPUT)
public SubscribableChannel send2();
}TestSender.java添加：
package com.example.mycloud.resource;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.messaging.Message;
import org.springframework.messaging.support.MessageBuilder;
import org.springframework.web.bind.annotation.RequestMapping;
import org.springframework.web.bind.annotation.RestController;
import com.example.mycloud.bean.Product;
@RestController
public class TestSender {
@Autowired
private ISendService sendService;
@RequestMapping("/send")
public void send() {
String msg = "abc..........";
Message message = MessageBuilder.withPayload(msg.getBytes()).build();
sendService.send().send(message);
}
//消息分组
@RequestMapping("/send2")
public void send2() {
Product product = new Product();
product.setId("1");
product.setName("abc");
Message message = MessageBuilder.withPayload(product).build();
sendService.send().send(message);
}
}配置文件添加：
spring:
cloud:
stream:
bindings:
groupOutput:
#指定输出通道对应的exchange主题名
destination: object-exchangestream-receiver模块
IReceiveService.java添加：
package com.example.mycloud.resource;
import org.springframework.cloud.stream.annotation.Input;
import org.springframework.messaging.SubscribableChannel;
/**
* 接收的接口
* @author user
*
*/
public interface IReceiveService {
@Input("stream-exchange")
public SubscribableChannel receive();
//消息分组
String INPUT = "groupInput";
@Input(IReceiveService.INPUT)
public SubscribableChannel receive2();
}ReceiveService.java添加：
package com.example.mycloud.service;
import org.springframework.cloud.stream.annotation.EnableBinding;
import org.springframework.cloud.stream.annotation.StreamListener;
import org.springframework.stereotype.Service;
import com.example.mycloud.bean.Product;
import com.example.mycloud.resource.IReceiveService;
@Service
@EnableBinding(IReceiveService.class) //绑定
public class ReceiveService {
@StreamListener("stream-exchange")
public void onReceive(byte[] msg) {
System.out.println("receive: " + new String(msg));
}
//消息分组
@StreamListener(IReceiveService.INPUT)
public void onReceive2(Product obj) {
System.out.println("receive: " + obj.toString());
}
}配置文件添加：
spring:
cloud:
stream:
bindings:
groupInput:
#指定输入通道对应的exchange主题名
destination: object-exchange
#具体分组对应MQ的队列名称，并持久化队列
group: groupQueue3、查看绑定情况
消息分组解决了，队列持久化问题。还有集群中一个队列对应多个consumer，只消费一次。
六、消息的分区
集群环境下，多个服务组成一个集群，如果想让相同的消息被同一个服务消费
1、stream-receiver模块，修改配置文件
spring:
cloud:
stream:
bindings:
partInput:
consumer:
#开启消费者分区功能
partitioned: true
#指定了当前消费者的总示例数量
instanceCount: 2
#设置当前实例的索引号，从0开始
instanceIndex: 02、stream-sender模块，修改配置文件
spring:
cloud:
stream:
bindings:
partOutput:
producer:
#通过该参数指定了分区键的表达式规则
partitionKeyExpression: payload
#指定了消息分区的数量
partitionCount: 2