1、启动集群

启动zookeeper,hadoop,saprk
# 1、三个节点
/usr/zookeeper/zookeeper-3.4.10/bin/zkServer.sh start 
/usr/zookeeper/zookeeper-3.4.10/bin/zkServer.sh status

# 2、master节点 启动hadoop
/usr/hadoop/hadoop-2.7.3/sbin/start-all.sh

# 3、master节点 启动spark
/usr/spark/spark-2.4.0-bin-hadoop2.7/sbin/start-all.sh

2、启动 kafka

# 在三个节点后台启动kafka
cd $KAFKA_HOME
./bin/kafka-server-start.sh config/server.properties &

cd $KAFKA_HOME
# 创建主题 badou_topic
./bin/kafka-topics.sh --create --zookeeper master:2181,slave1:2181,slave2:2181 --replication-factor 3 --partitions 6 --topic badou_topic

# 创建生产者producer
./bin/kafka-console-producer.sh --broker-list master:9092,slave1:9092,slave2:9092  --topic badou_topic

# 创建消费者consumer
./bin/kafka-console-consumer.sh --from-beginning  --topic badou_topic --bootstrap-server master:9092,slave1:9092,slave2:9092

3、使用IDEA消费kafka数据

对2s内，拉取的数据进行wordCount处理。

import org.apache.log4j.{Level, Logger}
import org.apache.spark.SparkConf
import org.apache.spark.streaming.kafka.KafkaUtils
import org.apache.spark.streaming.{Seconds, StreamingContext}

object ReceiverTest01 {
  def main(args: Array[String]): Unit = {
    val Array(group_id, topic, exectime) = Array("group_badou_topic", "badou_topic", "2")
    
    val conf = new SparkConf().setAppName("Receiver Test").setMaster("local[2]")
    Logger.getLogger("org.apache.spark").setLevel(Level.ERROR)
    val ssc = new StreamingContext(conf, Seconds(exectime.toInt))
    //  定义外部传入的topic， 可以指定多个topic
    val topicSet = topic.split(",").toSet
    //  定义线程
    val numThreads = 1
    val topicMap = topicSet.map((_, numThreads.toInt)).toMap
    val zkQuorum = "192.168.142.128:2181" // master的zookeeper地址
    
    //  createStream 表示receiver方式，核心操作
    val lines = KafkaUtils.createStream(ssc, zkQuorum, group_id, topicMap).map(_._2)
    lines.map((_,1L)).reduceByKey(_+_).print()
    //  (null,1) !==> map(_._2)
    ssc.start()
    ssc.awaitTermination()
  }
}

运行代码程序，在xshell的kafka–>producer，进行数据输入

例如输入：  代码结果会显示：
word
word
word
set
set
set
srt