Apache RocketMQ4.0.0

阿里rocketmq，有广泛的应用场景，订单，交易，充值，流计算，消息推送，日志流式处理，binglog分发等多个业务领域。

已正式成为apache项目，也于年前刚发布4.0.0版本，本文初试服务端和客户端

 

1 服务端安装

1.0 安装准备

CentOS release 6.8 (Final)

1. 64bit OS, Linux/Unix/Mac is recommended;
2. 64bit JDK 1.7+;
3. Maven 3.2.x
4. Git

 

1.1 编译

  > git clone https://github.com/apache/incubator-rocketmq.git
  > cd incubator-rocketmq
  > mvn clean package install -Prelease-all assembly:assembly -U
  > cd target/apache-rocketmq-all/

 

1.2 启动NameServer

 > nohup sh bin/mqnamesrv &
  > tail -f ~/logs/rocketmqlogs/namesrv.log
  The Name Server boot success...

 

1.3 启动Broker

  > nohup sh bin/mqbroker -n localhost:9876 &
  > tail -f ~/logs/rocketmqlogs/broker.log 
  The broker[%s, 172.30.30.233:10911] boot success...

 

 

2 客户端

2.1 客户示例客户端

运行生产者producer

 > export NAMESRV_ADDR=localhost:9876
 > sh bin/tools.sh org.apache.rocketmq.example.quickstart.Producer
 SendResult [sendStatus=SEND_OK, msgId= ...

运行消费者

 > sh bin/tools.sh org.apache.rocketmq.example.quickstart.Consumer
 ConsumeMessageThread_%d Receive New Messages: [MessageExt...

 

2.2 java客户端编写

pom.xmlsg

<rocketmq.version>4.0.0-incubating</rocketmq.version>

<dependency>
    <groupId>org.apache.rocketmq</groupId>
    <artifactId>rocketmq-client</artifactId>
    <version>${rocketmq.version}</version>
</dependency>
<dependency>
    <groupId>org.apache.rocketmq</groupId>
    <artifactId>rocketmq-all</artifactId>
    <version>${rocketmq.version}</version>
</dependency>

 

生产者

import org.apache.rocketmq.client.exception.MQClientException;
import org.apache.rocketmq.client.producer.DefaultMQProducer;
import org.apache.rocketmq.client.producer.SendResult;
import org.apache.rocketmq.common.message.Message;

import java.util.concurrent.TimeUnit;

public class Producer {
    public static void main(String[] args) throws MQClientException,
            InterruptedException {
        /**
         * 一个应用创建一个Producer，由应用来维护此对象，可以设置为全局对象或者单例<br>
         * 注意：ProducerGroupName需要由应用来保证唯一<br>
         * ProducerGroup这个概念发送普通的消息时，作用不大，但是发送分布式事务消息时，比较关键，
         * 因为服务器会回查这个Group下的任意一个Producer
         */
        DefaultMQProducer producer = new DefaultMQProducer("ProducerGroupName");
        producer.setNamesrvAddr("192.168.56.101:9876");
        producer.setInstanceName("Producer");
        producer.setVipChannelEnabled(false);

        /**
         * Producer对象在使用之前必须要调用start初始化，初始化一次即可<br>
         * 注意：切记不可以在每次发送消息时，都调用start方法
         */
        producer.start();

        /**
         * 下面这段代码表明一个Producer对象可以发送多个topic，多个tag的消息。
         * 注意：send方法是同步调用，只要不抛异常就标识成功。但是发送成功也可会有多种状态，<br>
         * 例如消息写入Master成功，但是Slave不成功，这种情况消息属于成功，但是对于个别应用如果对消息可靠性要求极高，<br>
         * 需要对这种情况做处理。另外，消息可能会存在发送失败的情况，失败重试由应用来处理。
         */
        for (int i = 0; i < 1; i++) {
            try {
                {
                    Message msg = new Message("TopicTest1",// topic
                            "TagA",// tag
                            "OrderID001",// key
                            ("Hello MetaQ").getBytes());// body
                    SendResult sendResult = producer.send(msg);
                    System.out.println(sendResult);
                }

                {
                    Message msg = new Message("TopicTest2",// topic
                            "TagB",// tag
                            "OrderID0034",// key
                            ("Hello MetaQ").getBytes());// body
                    SendResult sendResult = producer.send(msg);
                    System.out.println(sendResult);
                }

                {
                    Message msg = new Message("TopicTest3",// topic
                            "TagC",// tag
                            "OrderID061",// key
                            ("Hello MetaQ").getBytes());// body
                    SendResult sendResult = producer.send(msg);
                    System.out.println(sendResult);
                }
            } catch (Exception e) {
                e.printStackTrace();
            }
            TimeUnit.MILLISECONDS.sleep(1000);
        }

        /**
         * 应用退出时，要调用shutdown来清理资源，关闭网络连接，从MetaQ服务器上注销自己
         * 注意：我们建议应用在JBOSS、Tomcat等容器的退出钩子里调用shutdown方法
         */
        producer.shutdown();
    }
}

 

消费者

import org.apache.rocketmq.client.consumer.DefaultMQPushConsumer;
import org.apache.rocketmq.client.consumer.listener.ConsumeConcurrentlyContext;
import org.apache.rocketmq.client.consumer.listener.ConsumeConcurrentlyStatus;
import org.apache.rocketmq.client.consumer.listener.MessageListenerConcurrently;
import org.apache.rocketmq.client.exception.MQClientException;
import org.apache.rocketmq.common.message.MessageExt;

import java.util.List;

public class PushConsumer {

    /**
     * 当前例子是PushConsumer用法，使用方式给用户感觉是消息从RocketMQ服务器推到了应用客户端。<br>
     * 但是实际PushConsumer内部是使用长轮询Pull方式从MetaQ服务器拉消息，然后再回调用户Listener方法<br>
     */
    public static void main(String[] args) throws InterruptedException,
            MQClientException {
        /**
         * 一个应用创建一个Consumer，由应用来维护此对象，可以设置为全局对象或者单例<br>
         * 注意：ConsumerGroupName需要由应用来保证唯一
         */
        DefaultMQPushConsumer consumer = new DefaultMQPushConsumer(
                "ConsumerGroupName");
        consumer.setNamesrvAddr("192.168.56.101:9876");
        consumer.setInstanceName("Consumber");

        /**
         * 订阅指定topic下tags分别等于TagA或TagC或TagD
         */
        consumer.subscribe("TopicTest1", "TagA || TagC || TagD");
        /**
         * 订阅指定topic下所有消息<br>
         * 注意：一个consumer对象可以订阅多个topic
         */
        consumer.subscribe("TopicTest2", "*");

        consumer.registerMessageListener(new MessageListenerConcurrently() {

            /**
             * 默认msgs里只有一条消息，可以通过设置consumeMessageBatchMaxSize参数来批量接收消息
             */
            @Override
            public ConsumeConcurrentlyStatus consumeMessage(
                    List<MessageExt> msgs, ConsumeConcurrentlyContext context) {
                System.out.println(Thread.currentThread().getName()
                        + " Receive New Messages: " + msgs.size());

                MessageExt msg = msgs.get(0);
                if (msg.getTopic().equals("TopicTest1")) {
                    // 执行TopicTest1的消费逻辑
                    if (msg.getTags() != null && msg.getTags().equals("TagA")) {
                        // 执行TagA的消费
                        System.out.println(new String(msg.getBody()));
                    } else if (msg.getTags() != null
                            && msg.getTags().equals("TagC")) {
                        // 执行TagC的消费
                    } else if (msg.getTags() != null
                            && msg.getTags().equals("TagD")) {
                        // 执行TagD的消费
                    }
                } else if (msg.getTopic().equals("TopicTest2")) {
                    System.out.println(new String(msg.getBody()));
                }

                return ConsumeConcurrentlyStatus.CONSUME_SUCCESS;
            }
        });

        /**
         * Consumer对象在使用之前必须要调用start初始化，初始化一次即可<br>
         */
        consumer.start();

        System.out.println("Consumer Started.");
    }
}

 

安装中问题汇总：

> nohup sh bin/mqnamesrv &

先不使用以上命令，如果出错，不会有显示

使用

sh bin/mqnamesrv

出现以下问题

#
# There is insufficient memory for the Java Runtime Environment to continue.
# Native memory allocation (mmap) failed to map 805306368 bytes for committing reserved memory.
# Possible reasons:
# The system is out of physical RAM or swap space
# In 32 bit mode, the process size limit was hit
# Possible solutions:
# Reduce memory load on the system
# Increase physical memory or swap space
# Check if swap backing store is full
# Use 64 bit Java on a 64 bit OS
# Decrease Java heap size (-Xmx/-Xms)
# Decrease number of Java threads
# Decrease Java thread stack sizes (-Xss)
# Set larger code cache with -XX:ReservedCodeCacheSize=
# This output file may be truncated or incomplete.
#
# Out of Memory Error (os_linux.cpp:2627), pid=1864, tid=0x00007f94d8319700

原因：rocketmq默认jvm配置较高，导致内存不足

解决：

runserver.sh

改成

JAVA_OPT="${JAVA_OPT} -server -Xms256m -Xmx256m -Xmn128m -XX:PermSize=128m -XX:MaxPermSize=320m"

 

同理，运行其他命令也建议先使用sh来运行，解决方法如下

runbroker.sh

JAVA_OPT="${JAVA_OPT} -server -Xms8g -Xmx8g -Xmn4g"

改成：

JAVA_OPT="${JAVA_OPT} -server -Xms256m -Xmx256m -Xmn128m

 

tools.sh

JAVA_OPT="${JAVA_OPT} -server -Xms1g -Xmx1g -Xmn256m -XX:PermSize=128m -XX:MaxPermSize=128m"

改成

JAVA_OPT="${JAVA_OPT} -server -Xms256m -Xmx256m -Xmn128m -XX:PermSize=128m -XX:MaxPermSize=128m"

 

 

调试时问题汇总：

运行java客户端的Producer.java，出现

Caused by: org.apache.rocketmq.remoting.exception.RemotingConnectException: connect to <*****:10911> failed

经调试，

org.apache.rocketmq.client.impl.producer.DefaultMQProducerImpl

String brokerAddr = this.mQClientFactory.findBrokerAddressInPublish(mq.getBrokerName());

通过broderName获取到的ip不正确

解决：临时解决方案是断点在该代码下方，将brokerAddr设置为正确ip

 

reference:

https://rocketmq.incubator.apache.org/docs/quick-start/

http://www.tuicool.com/articles/mEBVneB

扩展阅读：

RocketMQ与Kafka对比（18项差异） http://blog.csdn.net/damacheng/article/details/42846549

Rocketmq源码分析（一）整体架构 https://yq.aliyun.com/articles/61141#

RocketMQ 源码分析 http://blog.csdn.net/vonzhoufz/article/details/47191883