1、RocketMQ概述
RocketMQ 是一款分布式、队列模型的消息中间件,具有以下特点: 能够保证严格的消息顺序 提供丰富的消息拉取模式 高效的订阅者水平扩展能力 实时的消息订阅机制 亿级消息堆积能力
2、RocketMQ包含的组件
NameServer:单点,供Producer和Consumer获取Broker地址
Producer:产生并发送消息
Consumer:接受并消费消息
Broker:消息暂存,消息转发
①首先生产者和消费者把自己的信息(包括地址信息)注册到nameServer上去
②broker有多个主,因为如果是一个主的话,那就是主从复制了,而多个主可以均摊消息存放。
Name Server
Name Server是RocketMQ的寻址服务。用于把Broker的路由信息做聚合。客户端依靠Name Server决定去获取对应topic的路由信息,从而决定对哪些Broker做连接。
Name Server是一个几乎无状态的结点,Name Server之间采取share-nothing的设计,互不通信。
对于一个Name Server集群列表,客户端连接Name Server的时候,只会选择随机连接一个结点,以做到负载均衡。
Name Server所有状态都从Broker上报而来,本身不存储任何状态,所有数据均在内存。
如果中途所有Name Server全都挂了,影响到路由信息的更新,不会影响和Broker的通信。
Broker
Broker是处理消息存储,转发等处理的服务器。
Broker以group分开,每个group只允许一个master,若干个slave。
只有master才能进行写入操作,slave不允许。
slave从master中同步数据。同步策略取决于master的配置,可以采用同步双写,异步复制两种。
客户端消费可以从master和slave消费。在默认情况下,消费者都从master消费,在master挂后,客户端由于从Name Server中感知到Broker挂机,就会从slave消费。
Broker向所有的NameServer结点建立长连接,注册Topic信息。
RocketMQ优点
1.强调集群无单点,可扩展
2.任意一点高可用,水平可扩展
3.海量消息堆积能力,消息堆积后,写入低延迟。
4.支持上万个队列
5.消息失败重试机制
6.消息可查询
7.开源社区活跃
8.成熟度(经过双十一考验)
RocketMQ环境安装
服务器配置
192.168.110.187 nameServer1,brokerServer1(意思就是说nameServer1,brokerServer1都在187这台服务器上)
192.168.110.188 nameServer2,brokerServer2
添加Host文件
vi /etc/hosts
192.168.110.187 rocketmq-nameserver1
192.168.110.187 rocketmq-master1
192.168.110.188 rocketmq-nameserver2
192.168.110.188 rocketmq-master2
service network restart
注意: Error:No suitable device found: no device found for connection “System eth0”
解决办法:
(1)ifconfig -a 查看物理 MAC HWADDR 的值
(2)vim 编辑文件 /etc/sysconfig/network-scripts/ifcfg-eth0中修改ifconfig中查出的MAC HWADDR值;
上传安装包
上传alibaba-rocketmq-3.2.6.tar.gz文件至/usr/local
tar -zxvf alibaba-rocketmq-3.2.6.tar.gz -C /usr/local
mv alibaba-rocketmq alibaba-rocketmq-3.2.6
ln -s alibaba-rocketmq-3.2.6 rocketmq(解释:创建软链接,什么是软链接呢?????)
创建存储路径【两台机器】
mkdir /usr/local/rocketmq/store
#mkdir /usr/local/rocketmq/store/commitlog
mkdir /usr/local/rocketmq/store/consumequeue
mkdir /usr/local/rocketmq/store/index
RocketMQ配置文件【两台机器】
vim /usr/local/rocketmq/conf/2m-noslave/broker-a.properties
vim /usr/local/rocketmq/conf/2m-noslave/broker-b.properties
修改日志配置文件【两台机器】
#mkdir -p /usr/local/rocketmq/logs
#cd /usr/local/rocketmq/conf && sed -i ‘s#${user.home}#/usr/local/rocketmq#g’ *.xml
修改启动NameServer【两台机器】
vim /usr/local/rocketmq/bin/runbroker.sh
JAVA_OPT="${JAVA_OPT} -server -Xms1g -Xmx1g -Xmn512m -
XX:PermSize=128m -XX:MaxPermSize=320m"
vim /usr/local/rocketmq/bin/runserver.sh
JAVA_OPT="${JAVA_OPT} -server -Xms1g -Xmx1g -Xmn512m -
XX:PermSize=128m -XX:MaxPermSize=320m"
启动NameServer【两台机器】
#cd /usr/local/rocketmq/bin
nohup sh mqnamesrv &
启动BrokerServer A
cd /usr/local/rocketmq/bin
#nohup sh mqbroker -c /usr/local/rocketmq/conf/2m-noslave/broker-a.properties >/dev/null 2>&1 &
netstat -ntlp
jps
tail -f -n 500 /usr/local/rocketmq/logs/rocketmqlogs/broker.log
tail -f -n 500 /usr/local/rocketmq/logs/rocketmqlogs/namesrv.log
启动BrokerServer B
cd /usr/local/rocketmq/bin
nohup sh mqbroker -c /usr/local/rocketmq/conf/2m-noslave/broker-b.properties >/dev/null 2>&1 &
netstat -ntlp
jps
tail -f -n 500 /usr/local/rocketmq/logs/rocketmqlogs/broker.log
tail -f -n 500 /usr/local/rocketmq/logs/rocketmqlogs/namesrv.log
RocketMQ Console
将rocketmq-web-console 部署到webapps目录中。
/usr/local/apache-tomcat-7.0.65/webapps/rocketmq-web-console/WEB-INF/classes/
修改config.properties
rocketmq.namesrv.addr=192.168.110.195:9876;192.168.110.199:9876
运行效果
Java操作RocketMQ
Pom
<dependencies>
<dependency>
<groupId>com.alibaba.rocketmq</groupId>
<artifactId>rocketmq-client</artifactId>
<version>3.0.10</version>
</dependency>
<dependency>
<groupId>com.alibaba.rocketmq</groupId>
<artifactId>rocketmq-all</artifactId>
<version>3.0.10</version>
<type>pom</type>
</dependency>
<dependency>
<groupId>ch.qos.logback</groupId>
<artifactId>logback-classic</artifactId>
<version>1.1.1</version>
</dependency>
<dependency>
<groupId>ch.qos.logback</groupId>
<artifactId>logback-core</artifactId>
<version>1.1.1</version>
</dependency>
<dependency>
<groupId>junit</groupId>
<artifactId>junit</artifactId>
<version>4.10</version>
<scope>test</scope>
</dependency>
</dependencies>
生产者
public class Producer {
public static void main(String[] args) throws MQClientException {
DefaultMQProducer producer = new DefaultMQProducer("rmq-group");
producer.setNamesrvAddr("192.168.110.195:9876;192.168.110.199:9876");
producer.setInstanceName("producer");
producer.start();
try {
for (int i = 0; i < 10; i++) {
Thread.sleep(1000); // 每秒发送一次MQ
Message msg = new Message("itmayiedu-topic", // topic 主题名称
"TagA", // tag 临时值
("itmayiedu-"+i).getBytes()// body 内容
);
SendResult sendResult = producer.send(msg);
System.out.println(sendResult.toString());
}
} catch (Exception e) {
e.printStackTrace();
}
producer.shutdown();
}
}
消费者
public class Consumer {
public static void main(String[] args) throws MQClientException {
DefaultMQPushConsumer consumer = new DefaultMQPushConsumer("rmq-group");
consumer.setNamesrvAddr("192.168.110.195:9876;192.168.110.199:9876");
consumer.setInstanceName("consumer");
consumer.subscribe("itmayiedu-topic", "TagA");
consumer.registerMessageListener(new MessageListenerConcurrently() {
@Override
public ConsumeConcurrentlyStatus consumeMessage(List<MessageExt> msgs, ConsumeConcurrentlyContext context) {
for (MessageExt msg : msgs) {
System.out.println(msg.getMsgId()+"---"+new String(msg.getBody()));
}
return ConsumeConcurrentlyStatus.CONSUME_SUCCESS;
}
});
consumer.start();
System.out.println("Consumer Started.");
}
}
RocketMQ重试机制
RocketMQ重试机制
MQ 消费者的消费逻辑失败时,可以通过设置返回状态达到消息重试的结果。
MQ 消息重试只针对集群消费方式生效;广播方式不提供失败重试特性,即消费失败后,失败消息不再重试,继续消费新的消息。
public class Consumer {
public static void main(String[] args) throws MQClientException {
DefaultMQPushConsumer consumer = new DefaultMQPushConsumer("rmq-group");
consumer.setNamesrvAddr("192.168.110.195:9876;192.168.110.199:9876");
consumer.setInstanceName("consumer");
consumer.subscribe("itmayiedu-topic", "TagA");
consumer.registerMessageListener(new MessageListenerConcurrently() {
@Override
public ConsumeConcurrentlyStatus consumeMessage(List<MessageExt> msgs, ConsumeConcurrentlyContext context) {
for (MessageExt msg : msgs) {
System.out.println(msg.getMsgId() + "---" + new String(msg.getBody()));
}
try {
int i = 1 / 0;
} catch (Exception e) {
e.printStackTrace();
// 需要重试:如果出异常了,肯定要重试
return ConsumeConcurrentlyStatus.RECONSUME_LATER;
}
// 不需要重试
return ConsumeConcurrentlyStatus.CONSUME_SUCCESS;
}
});
consumer.start();
System.out.println("Consumer Started.");
}
}
注意:每次重试后,消息ID都不一致,所以不能使用消息ID判断幂等。
RocketMQ如何解决消息幂等
注意:每次重试后,消息ID都不一致,所以不能使用消息ID判断幂等。
解决办法:使用自定义全局ID判断幂等,例如流水ID、订单号
使用msg.setKeys 进行区分
生产者:
public class Producer {
public static void main(String[] args) throws MQClientException {
DefaultMQProducer producer = new DefaultMQProducer("rmq-group");
producer.setNamesrvAddr("192.168.110.195:9876;192.168.110.199:9876");
producer.setInstanceName("producer");
producer.start();
try {
for (int i = 0; i < 1; i++) {
Thread.sleep(1000); // 每秒发送一次MQ
Message msg = new Message("itmayiedu-topic", // topic 主题名称
"TagA", // tag 临时值
("itmayiedu-6" + i).getBytes()// body 内容
);
msg.setKeys(System.currentTimeMillis() + "");
SendResult sendResult = producer.send(msg);
System.out.println(sendResult.toString());
}
} catch (Exception e) {
e.printStackTrace();
}
producer.shutdown();
}
}
消费者:
static private Map<String, String> logMap = new HashMap<>();
public static void main(String[] args) throws MQClientException {
DefaultMQPushConsumer consumer = new DefaultMQPushConsumer("rmq-group");
consumer.setNamesrvAddr("192.168.110.195:9876;192.168.110.199:9876");
consumer.setInstanceName("consumer");
consumer.subscribe("itmayiedu-topic", "TagA");
consumer.registerMessageListener(new MessageListenerConcurrently() {
@Override
public ConsumeConcurrentlyStatus consumeMessage(List<MessageExt> msgs, ConsumeConcurrentlyContext context) {
String key = null;
String msgId = null;
try {
for (MessageExt msg : msgs) {
key = msg.getKeys();
③ if (logMap.containsKey(key)) {
// 无需继续重试。
System.out.println("key:"+key+",无需重试...");
return ConsumeConcurrentlyStatus.CONSUME_SUCCESS;
}
msgId = msg.getMsgId();
System.out.println("key:" + key + ",msgid:" + msgId + "---" + new String(msg.getBody()));
int i = 1 / 0;//模拟长时间消费者未返回success,那么就会重试机制,这个时候logMap里面已经存储了该消息的全局id,所以③判断为true,然后return success
}
} catch (Exception e) {
e.printStackTrace();
return ConsumeConcurrentlyStatus.RECONSUME_LATER;
} finally {
logMap.put(key, msgId);不管成不成功都要把全局id存储在logMap中
}
return ConsumeConcurrentlyStatus.CONSUME_SUCCESS;
}
});
consumer.start();
System.out.println("Consumer Started.");
}
解析:为什么会有重试机制,就是因为消费者迟迟不给mq返回success,这才是根本原因。