kafka使用实例

定义一个procucer

package cn.vko.common.kafka;

import java.util.Properties;

import org.apache.commons.lang3.StringUtils;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

import cn.vko.common.utils.mybatis.GenCreateInterceptor;

import kafka.javaapi.producer.Producer;
import kafka.producer.KeyedMessage;
import kafka.producer.ProducerConfig;

public class VkoProducer {
//    private Logger log = LoggerFactory.getLogger(VkoProducer.class);
    private String metadataBrokerList;
    private Producer<String, String> producer;

    public VkoProducer(String metadataBrokerList) {
        super();
        if(StringUtils.isEmpty(metadataBrokerList)){
            String message = "metadataBrokerList 不可以为空";
//            log.error(message);
            throw new RuntimeException(message);
        }
        this.metadataBrokerList = metadataBrokerList;
        // 设置配置属性
        Properties props = new Properties();
        props.put("metadata.broker.list", metadataBrokerList);
        props.put("serializer.class", "kafka.serializer.StringEncoder");
        props.put("key.serializer.class", "kafka.serializer.StringEncoder");
        props.put("request.required.acks", "1");        
        //props.put("producer.type", "async");
        props.put("queue.buffering.max.ms", "5000");
        props.put("queue.buffering.max.messages", "30000");
        props.put("queue.enqueue.timeout.ms", "-1");
        props.put("batch.num.messages", "1");
        // 可选配置，如果不配置，则使用默认的partitioner
        //props.put("partitioner.class", "cn.vko.kafka.PartitionerDemo");
        // 触发acknowledgement机制，否则是fire and forget，可能会引起数据丢失
        // 值为0,1,-1,可以参考
        // http://kafka.apache.org/08/configuration.html
        ProducerConfig config = new ProducerConfig(props);
        producer = new Producer<String, String>(config);
    }

    /**
     * 单条插入队列
     * @param msg
     * @param topic 主题
     * @return
     */
    public String send(String topic, String msg) {
//        Long start = System.currentTimeMillis();
        KeyedMessage<String, String> data = new KeyedMessage<String, String>(topic, msg);
        producer.send(data);
//        log.info("发送消息耗时：{}",System.currentTimeMillis()- start);
        return "ok";
    }
}

定义一个receiver

package cn.vko.common.kafka;

import java.util.HashMap;
import java.util.List;
import java.util.Map;
import java.util.Properties;

import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

import cn.vko.common.utils.mybatis.GenCreateInterceptor;
import cn.vko.component.pageframework.util.StringUtil;

import kafka.consumer.Consumer;
import kafka.consumer.ConsumerConfig;
import kafka.consumer.ConsumerIterator;
import kafka.consumer.KafkaStream;
import kafka.javaapi.consumer.ConsumerConnector;

public class Receiver {
    private Logger log = LoggerFactory.getLogger(Receiver.class);
    private String zookeeperConnect;
    private String groupId;
    private String topic;
    private VkoConsumer vkoConsumer;
    
    /**
     * 创建收件人
     * @param zookeeperConnect zk集群地址，逗号分隔
     * @param groupId 组id
     * @param topic 主题
     * @param vkoConsumer 处理器
     */
    public Receiver(String zookeeperConnect, String groupId, String topic,VkoConsumer vkoConsumer) {
        super();
        if(StringUtil.isEmpty(zookeeperConnect)){
            String message = "zookeeperConnect 不可以为空";
            log.error(message);
            throw new RuntimeException(message);
        }
        if(StringUtil.isEmpty(groupId)){
            String message = "groupId 不可以为空";
            log.error(message);
            throw new RuntimeException(message);
        }
        if(StringUtil.isEmpty(topic)){
            String message = "topic 不可以为空";
            log.error(message);
            throw new RuntimeException(message);
        }
        if(vkoConsumer == null){
            String message = "vkoConsumer 不可以为空";
            log.error(message);
            throw new RuntimeException(message);
        }
        this.zookeeperConnect = zookeeperConnect;
        this.groupId = groupId;
        this.topic = topic;
        this.vkoConsumer = vkoConsumer;
        log.info("kafka vkoConsumer 创建完成：groupId：{},topic:{},zookeeperConnect:{}",groupId, topic, zookeeperConnect);
        receive();
    }

    private void receive(){
        Properties props = new Properties();
        props.put("zookeeper.connect", zookeeperConnect);
        props.put("group.id", groupId);
        props.put("zookeeper.session.timeout.ms", "14000");
        props.put("zookeeper.sync.time.ms", "200");
        props.put("auto.commit.interval.ms", "1000");
        ConsumerConfig conf = new ConsumerConfig(props);
        ConsumerConnector cc = Consumer.createJavaConsumerConnector(conf);
         Map<String, Integer> topicCountMap = new HashMap<String, Integer>();
         // 目前每个topic都是2个分区
         topicCountMap.put(topic,2);
         Map<String, List<KafkaStream<byte[], byte[]>>> consumerMap = cc.createMessageStreams(topicCountMap);
            List<KafkaStream<byte[], byte[]>> streams = consumerMap.get(topic);
            for (final KafkaStream<byte[], byte[]> stream : streams) {
                new Thread(){
                    public void run(){
                        ConsumerIterator<byte[], byte[]> it = stream.iterator();
                        while(it.hasNext()){
                            String msg = new String(it.next().message());
                            try{
                             vkoConsumer.dealMsg(msg);
                            }catch(Exception e){
                                log.error("kafka vkoConsumer topic:{} 收到消息：{} 消费异常 xxxxxxxxxxxxxxxxxx", topic, msg,e);
                            }
                            log.info("kafka vkoConsumer topic:{} 收到消息：{}", topic, msg);
                        }
                    }
                }.start();
                log.info("kafka vkoConsumer 启动完成：groupId：{},topic:{},zookeeperConnect:{}",groupId, topic, zookeeperConnect);
            }
            log.info("kafka vkoConsumer 准备接收消息：groupId：{},topic:{},zookeeperConnect:{}",groupId, topic, zookeeperConnect);
    }
}

package cn.vko.common.kafka;

public interface VkoConsumer {
    public void dealMsg(String strings);
}

在需要consumer的程序中定义一个消费实现类,并注入到receiver中,这样spring容器启动时会自动创建一个receiver,进行对特定的topic消费

<!-- 定义消息处理器 -->
    <bean id="sellStatConsumer" class="cn.vko.sell.kafka.consumer.SellStatConsumer" ></bean>

    <!-- 定义收信人 receiver -->
      <bean id="sellStatReceiver" class="cn.vko.common.kafka.Receiver">

         <constructor-arg index="0" value="${zookeeper.connect}" /><!-- _zookeeper集群地址,如: zkserver1.vko.cn:2181,zkserver2.vko.cn:2181,zkserver3.vko.cn:2181_-->

         <constructor-arg index="1" value="${sellstat.group.id}" /><!-- _消费者所属组id字符串 ，如:vko_group_article_read_count_-->

         <constructor-arg index="2" value="${kafka.sellstat.topics}"/><!-- _要消费的消息主题，如：vko_group_-->

         <constructor-arg index="3" ref="sellStatConsumer" /> <!--_上面定义的消息处理器_-->
    </bean>