java-RocketMQ

RocketMQ单机支持1万以上的持久化队列，前提是足够的内存、硬盘空间，过期数据数据删除（RocketMQ中的消息队列长度不是无限的，只是足够大的内存+数据定时删除）

RocketMQ版本：3.1.4

 

一，部署NameServer：

1，安装JDK并设置JAVA_HOME环境变量（启动脚本依赖JAVA_HOME环境变量）

2，cd /alibaba-rocketmq/bin进入RocketMQ的bin目录

2，调用nohup sh mqnamesrv &启动NameServer

报错如下：

 : command not found
 : command not found
 mqnamesrv: line 35: syntax error: unexpected end of file

在bin目录下调用dos2unix *将所有文件转化为unix格式，再次调用nohup sh mqnamesrv &

报错如下：

 /home/hadoop/alibaba-rocketmq
 Invalid initial heap size: -Xms4g
 The specified size exceeds the maximum representable size.
 Could not create the Java virtual machine.

由于安装的JDK版本为32位，4g超过了JDK所支持的最大内存，不过32位JDK也无法发挥出RocketMQ的优势，换成64位JDK

这次启动成功

 [hadoop@hadoop bin]$ nohup sh mqnamesrv &
 [1] 17676
 [hadoop@hadoop bin]$ nohup: appending output to “nohup.out”

 [hadoop@hadoop bin]$ cat nohup.out
 The Name Server boot success.
 [hadoop@hadoop bin]$ jps
 17682 NamesrvStartup
 17800 Jps

NameServer监听端口：9876

 nettyServerConfig.setListenPort(9876);

如果服务器内存不够，可以修改runserver.sh脚本（mqnamesrv文件中通过runserver.sh脚本调用Name Server的主函数com.alibaba.rocketmq.namesrv.NamesrvStartup启动Name Server）中的JAVA_OPT_1参数

 JAVA_OPT_1="-server -Xms4g -Xmx4g -Xmn2g -XX:PermSize=128m -XX:MaxPermSize=320m"

二，部署Broker：消息中转角色，负责存储消息，转发消息

Broker集群有多种配置方式：

1，单Master

    优点：除了配置简单没什么优点

    缺点：不可靠，该机器重启或宕机，将导致整个服务不可用

2，多Master

    优点：配置简单，性能最高

    缺点：可能会有少量消息丢失（配置相关），单台机器重启或宕机期间，该机器下未被消费的消息在机器恢复前不可订阅，影响消息实时性

3，多Master多Slave，每个Master配一个Slave，有多对Master-Slave，集群采用异步复制方式，主备有短暂消息延迟，毫秒级

    优点：性能同多Master几乎一样，实时性高，主备间切换对应用透明，不需人工干预

    缺点：Master宕机或磁盘损坏时会有少量消息丢失

4，多Master多Slave，每个Master配一个Slave，有多对Master-Slave，集群采用同步双写方式，主备都写成功，向应用返回成功

    优点：服务可用性与数据可用性非常高

    缺点：性能比异步集群略低，当前版本主宕备不能自动切换为主

Master和Slave的配置文件参考conf目录下的配置文件

Master与Slave通过指定相同的brokerName参数来配对，Master的BrokerId必须是0，Slave的BrokerId必须是大于0的数

一个Master下面可以挂载多个Slave，同一Master下的多个Slave通过指定不同的BrokerId来区分

部署一Master一Slave，集群采用异步复制方式：

Master：

 [hadoop@hadoop bin]$ nohup sh mqbroker -n "192.168.58.163:9876" -c ../conf/2m-2s-async/broker-a.properties &
 [2] 25493
 [hadoop@hadoop bin]$ nohup: appending output to “nohup.out”

 [hadoop@hadoop bin]$ cat nohup.out
 load config properties file OK, ../conf/2m-2s-async/broker-a.properties
 The broker[broker-a, 192.168.58.163:10911] boot success. and name server is 192.168.58.163:9876
 [hadoop@hadoop bin]$ jps
 25500 BrokerStartup
 25545 Jps
 17682 NamesrvStartup

Slave：

 [hadoop@hadoop bin]$ nohup sh mqbroker -n "192.168.58.163:9876" -c ../conf/2m-2s-async/broker-a-s.properties &
 [1] 1974
 [hadoop@hadoop bin]$ nohup: appending output to “nohup.out”

 [hadoop@hadoop bin]$ cat nohup.out
 load config properties file OK, ../conf/2m-2s-async/broker-a-s.properties
 The broker[broker-a, 192.168.58.164:10911] boot success. and name server is 192.168.58.163:9876
 [hadoop@hadoop bin]$ jps
 2071 Jps
 1981 BrokerStartup

Broker监听端口：10911

 nettyServerConfig.setListenPort(10911);

如果服务器内存不够，可以修改runbroker.sh脚本（mqbroker文件中通过runbroker.sh脚本调用Broker的主函数com.alibaba.rocketmq.broker.BrokerStartup启动Broker）的JAVA_OPT_1参数

 JAVA_OPT_1="-server -Xms4g -Xmx4g -Xmn2g -XX:PermSize=128m -XX:MaxPermSize=320m"

三，Producer

必须要设置Name Server地址

 package com.sean;

 import com.alibaba.rocketmq.client.producer.DefaultMQProducer;
 import com.alibaba.rocketmq.client.producer.SendResult;
 import com.alibaba.rocketmq.common.message.Message;

 public class Producer {
     public static void main(String[] args){
         DefaultMQProducer producer = new DefaultMQProducer("Producer");
         producer.setNamesrvAddr("192.168.58.163:9876");
         try {
             producer.start();

             Message msg = new Message("PushTopic",
                     "push",
                     "1",
                     "Just for test.".getBytes());

             SendResult result = producer.send(msg);
             System.out.println("id:" + result.getMsgId() +
                     " result:" + result.getSendStatus());

             msg = new Message("PushTopic",
                     "push",
                     "2",
                     "Just for test.".getBytes());

             result = producer.send(msg);
             System.out.println("id:" + result.getMsgId() +
                     " result:" + result.getSendStatus());

             msg = new Message("PullTopic",
                     "pull",
                     "1",
                     "Just for test.".getBytes());

             result = producer.send(msg);
             System.out.println("id:" + result.getMsgId() +
                     " result:" + result.getSendStatus());
         } catch (Exception e) {
             e.printStackTrace();
         }finally{
             producer.shutdown();
         }
     }
 }

四，Consumer

必须要设置Name Server地址

 package com.sean;

 import java.util.List;

 import com.alibaba.rocketmq.client.consumer.DefaultMQPushConsumer;
 import com.alibaba.rocketmq.client.consumer.listener.ConsumeConcurrentlyContext;
 import com.alibaba.rocketmq.client.consumer.listener.ConsumeConcurrentlyStatus;
 import com.alibaba.rocketmq.client.consumer.listener.MessageListenerConcurrently;
 import com.alibaba.rocketmq.common.consumer.ConsumeFromWhere;
 import com.alibaba.rocketmq.common.message.Message;
 import com.alibaba.rocketmq.common.message.MessageExt;

 public class Consumer {
     public static void main(String[] args){
         DefaultMQPushConsumer consumer =
                 new DefaultMQPushConsumer("PushConsumer");
         consumer.setNamesrvAddr("192.168.58.163:9876");
         try {
             //订阅PushTopic下Tag为push的消息
             consumer.subscribe("PushTopic", "push");
             //程序第一次启动从消息队列头取数据
             consumer.setConsumeFromWhere(
                     ConsumeFromWhere.CONSUME_FROM_FIRST_OFFSET);
             consumer.registerMessageListener(
                 new MessageListenerConcurrently() {
                     public ConsumeConcurrentlyStatus consumeMessage(
                             List<MessageExt> list,
                             ConsumeConcurrentlyContext Context) {
                         Message msg = list.get(0);
                         System.out.println(msg.toString());
                         return ConsumeConcurrentlyStatus.CONSUME_SUCCESS;
                     }
                 }
             );
             consumer.start();
         } catch (Exception e) {
             e.printStackTrace();
         }
     }
 }

先运行Consumer，然后运行Producer

Producer运行结果：

 id:C0A83AA300002A9F00000000000009EA result:SEND_OK
 id:C0A83AA300002A9F0000000000000A77 result:SEND_OK
 id:C0A83AA300002A9F0000000000000B04 result:SEND_OK

Consumer运行结果：

 MessageExt [queueId=1, storeSize=141, queueOffset=6, sysFlag=0, bornTimestamp=1403765668792, bornHost=/192.168.31.130:60985, storeTimestamp=1403765527374, storeHost=/192.168.58.163:10911, msgId=C0A83AA300002A9F0000000000000A77, commitLogOffset=2679, bodyCRC=753746584, reconsumeTimes=0, preparedTransactionOffset=0, toString()=Message [topic=PushTopic, flag=0, properties={TAGS=push, KEYS=2, WAIT=true, MAX_OFFSET=7, MIN_OFFSET=0}, body=14]]
 MessageExt [queueId=0, storeSize=141, queueOffset=6, sysFlag=0, bornTimestamp=1403765668698, bornHost=/192.168.31.130:60985, storeTimestamp=1403765527356, storeHost=/192.168.58.163:10911, msgId=C0A83AA300002A9F00000000000009EA, commitLogOffset=2538, bodyCRC=753746584, reconsumeTimes=0, preparedTransactionOffset=0, toString()=Message [topic=PushTopic, flag=0, properties={TAGS=push, KEYS=1, WAIT=true, MAX_OFFSET=7, MIN_OFFSET=0}, body=14]]