zookeeper+Kafka集群部署

一 准备工作

准备3台机器，IP地址分别为：192.168.46.130（131，132）

下载 jdk-8u121-linux-x64.tar.gz   zookeeper-3.4.6.tar.gz   kafka_2.11-0.10.1.0.tgz

1.修改主机名 /etc/hosts 及/etc/sysconfig/network

hosts:

192.168.46.130 kafka1
192.168.46.131 kafka2
192.168.46.132 kafka3

network

分别将network中的主机名修改为 kafka1 kafka2 kafka3

2.jdk安装配置省略

3.zookeeper部署

1) 解压zookeeper-3.4.6.tar.gz

2) tar -zxvf zookeeper-3.4.6.tar.gz

3) 建立myid  在/usr/local/目录下建立 zkData目录,即/usr/local/zkData,

4) 并执行  echo 1 >myid (kafka1服务器上)  echo 2 >myid (kafka2服务器上)  echo 3 >myid (kafka3服务器上) 

5) 三台服务器的zoo.cfg配置文件为

zoo.cfg

tickTime=2000
initLimit=10
syncLimit=5
dataDir=/usr/local/zkData
clientPort=2181
server.1=192.168.46.130:2888:3888
server.2=192.168.46.131:2888:3888
server.3=192.168.46.132:2888:3888

4.kafka部署(解压kafka_2.11-0.10.1.0.tgz文件)

1).进入config目录，修改server.properties（逐个机器修改）

broker.id=1 (其他服务器2/3)

zookeeper.connect=192.168.46.130:2181,192.168.46.131:2181,192.168.46.132:2181

zookeeper.connection.timeout.ms=6000

port = 9092
host.name = 192.168.46.131

5.启动每台服务器上的zookeeper

1) 启动前关闭各台服务器的防火墙:service iptables stop 

2) bin/zkServer.sh start 启动

3) bin/zkServer.sh status 查看状态

6 启动每台服务器的kafka：

        > bin/kafka-server-start.sh config/server.properties &

7.集群测试

    1.创建一个topic

        > bin/kafka-topics.sh --create --zookeeper 192.168.46.130:2181 --replication-factor 3 --partitions 1

            --topic test-topic

 

    2.查看创建的topic

        > bin/kafka-topics.sh --describe --zookeeper 192.168.46.130:2181 --topic test-topic

            Topic:test-replicated-topic    PartitionCount:1    ReplicationFactor:3    Configs:

            Topic: test-replicated-topic    Partition: 0    Leader: 1    Replicas: 1,2,0    Isr: 1,2,0

 

     3.查看topic列表

          > bin/kafka-topics.sh --list --zookeeper 192.168.46.130 :2181 

　　　　test-topic

 　　 查看列表及具体信息

          > bin/kafka-topics.sh --zookeeper localhost --describe

 

    4.查看集群情况：

        >bin/kafka-topics.sh --describe --zookeeper 192.168.46.131:2181 --topic test-topic

        >bin/kafka-topics.sh --describe --zookeeper 192.168.46.132:2181 --topic test-topic

        发现都能看到test-topic。

 

     5.生产消息(定义不同的生产者)

          > bin/kafka-console-producer.sh --broker-list 192.168.46.130:9092 -topic test-topic

         接着输入生产的内容:如 haha

     6.消费消息(在不同的服务器上执行下面的命令后都可接受到生产的消息)

          > bin/kafka-console-consumer.sh --zookeeper 192.168.46.131:2181 --from-beginning --topic test-topic

          > bin/kafka-console-consumer.sh --zookeeper 192.168.46.132:2181 --from-beginning --topic test-topic

         消费端的内容:haha

       每个节点既可以作为生产者也可以作为消费者,一个节点要么为生产者要么为消费者

 8 客户端调用(参考:http://blog.csdn.net/u011622226/article/details/53520382)

maven 引入 

<dependency>
 <groupId>org.apache.kafkagroupId>
 <artifactId>kafka_2.11artifactId>
 <version>0.10.1.0version>
dependency>

package com.fuliwd.kafka;

import java.util.Properties;
import org.apache.kafka.clients.producer.Callback;
import org.apache.kafka.clients.producer.KafkaProducer;
import org.apache.kafka.clients.producer.ProducerRecord;
import org.apache.kafka.clients.producer.RecordMetadata;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

public class KafkaProducerService {
    private static Logger LOG = LoggerFactory
            .getLogger(KafkaProducerService.class);

    public static void main(String[] args) {
        Properties props = new Properties();
        //zookeeper地址
        props.put("bootstrap.servers", "192.168.46.130:9092,192.168.46.131:9092,192.168.46.132:9092");
        props.put("retries", 3);
        props.put("linger.ms", 1);
        props.put("key.serializer",
                "org.apache.kafka.common.serialization.StringSerializer");
        props.put("value.serializer",
                "org.apache.kafka.common.serialization.StringSerializer");
        KafkaProducer producer = new KafkaProducer(
                props);
        //test-topic为队列名称
        for (int i = 0; i < 1; i++) {
            ProducerRecord record = new ProducerRecord(
                    "test-topic", "11", "客户端调用 yoyo=======>" + i);
            producer.send(record, new Callback() {
                public void onCompletion(RecordMetadata metadata, Exception e) {
                    // TODO Auto-generated method stub
                    if (e != null)
                        System.out.println("the producer has a error:"
                                + e.getMessage());
                    else {
                        System.out
                                .println("The offset of the record we just sent is: "
                                        + metadata.offset());
                        System.out
                                .println("The partition of the record we just sent is: "
                                        + metadata.partition());
                    }
                }
            });
            try {
                Thread.sleep(1000);
                // producer.close();
            } catch (InterruptedException e1) {
                // TODO Auto-generated catch block
                e1.printStackTrace();
            }
        }
    }
}

package com.fuliwd.kafka;

import java.util.ArrayList;
import java.util.Arrays;
import java.util.List;
import java.util.Properties;
import org.apache.kafka.clients.consumer.ConsumerRecord;
import org.apache.kafka.clients.consumer.ConsumerRecords;
import org.apache.kafka.clients.consumer.KafkaConsumer;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

/**
 * @author Joker
 * 自己控制偏移量提交
 * 很多时候，我们是希望在获得消息并经过一些逻辑处理后，才认为该消息已被消费，这可以通过自己控制偏移量提交来实现。
 */
public class ManualOffsetConsumer {
    private static Logger LOG = LoggerFactory.getLogger(ManualOffsetConsumer.class);

    public static void main(String[] args) {
        // TODO Auto-generated method stub
        Properties props = new Properties();
        //设置brokerServer(kafka)ip地址 zookeeper地址
        props.put("bootstrap.servers", "192.168.46.130:9092,192.168.46.131:9092,192.168.46.132:9092");
        //设置consumer group name
        props.put("group.id","mygroup11");
        props.put("enable.auto.commit", "false");
        //设置使用最开始的offset偏移量为该group.id的最早。如果不设置，则会是latest即该topic最新一个消息的offset
        //如果采用latest，消费者只能得道其启动后，生产者生产的消息
        props.put("auto.offset.reset", "earliest");
        //设置心跳时间
        props.put("session.timeout.ms", "30000");
        props.put("key.deserializer", "org.apache.kafka.common.serialization.StringDeserializer");
        props.put("value.deserializer", "org.apache.kafka.common.serialization.StringDeserializer");
        KafkaConsumer consumer = new KafkaConsumer(props);
        consumer.subscribe(Arrays.asList("test-topic"));//队列名称
        final int minBatchSize = 5;  //批量提交数量
        List> buffer = new ArrayList>();
        while (true) {
            ConsumerRecords records = consumer.poll(100);
            for (ConsumerRecord record : records) {
                System.out.println("consumer message values is "+record.value()+" and the offset is "+ record.offset());
                buffer.add(record);
            }
            if (buffer.size() >= minBatchSize) {
                System.out.println("now commit offset"+buffer.size());
                consumer.commitSync();
                buffer.clear();
            }
        }
    }
}

下图为服务器端进行消息生产和客户端调用生产的执行情况(这里只讲kafka1设置为生产者,kafka2为消费者,也可以将kafka3设置生产或消费者)