StructuredStreaming&Kafka中的实时ETL案例及常见问题

实时ETL

在 java 项目中，导入 jar 包：

java 复制代码

<dependencies>
  <dependency>
    <groupId>org.apache.kafka</groupId>
    <artifactId>kafka-clients</artifactId>
    <version>3.0.0</version>
  </dependency>

  <dependency>
    <groupId>org.slf4j</groupId>
    <artifactId>slf4j-log4j12</artifactId>
    <version>1.7.25</version>
  </dependency>

  <dependency>
    <groupId>org.projectlombok</groupId>
    <artifactId>lombok</artifactId>
    <version>1.18.24</version>
  </dependency>

  <dependency>
    <groupId>com.alibaba</groupId>
    <artifactId>fastjson</artifactId>
    <version>2.0.50</version>
  </dependency>

java 复制代码

package com.bigdata.moni;


import lombok.AllArgsConstructor;
import lombok.Data;
import lombok.NoArgsConstructor;

@Data
@NoArgsConstructor
@AllArgsConstructor
public class StationLog {

        private String stationId;
        private String callOut;
        private String callIn;
        private String callStatus;
        private long callTime;
        private int duration;

}

java 复制代码

package com.bigdata.smartedu;

import com.alibaba.fastjson.JSON;
import com.bigdata.moni.StationLog;
import org.apache.kafka.clients.producer.KafkaProducer;
import org.apache.kafka.clients.producer.ProducerConfig;
import org.apache.kafka.clients.producer.ProducerRecord;

import java.util.Properties;
import java.util.Random;

public class SendStationLogProducer {

    public static void main(String[] args) throws InterruptedException {
        Properties properties = new Properties();

        properties.put(ProducerConfig.KEY_SERIALIZER_CLASS_CONFIG,"org.apache.kafka.common.serialization.StringSerializer");
        properties.put(ProducerConfig.VALUE_SERIALIZER_CLASS_CONFIG,"org.apache.kafka.common.serialization.StringSerializer");
        properties.put(ProducerConfig.BOOTSTRAP_SERVERS_CONFIG,"bigdata01:9092");

        String[] arr = {"fail", "busy", "barring", "success", "success", "success",
                "success", "success", "success", "success", "success", "success"};
        Random random = new Random();
        KafkaProducer<String, String> kafkaProducer = new KafkaProducer<String, String>(properties);
        while(true){

            String callOut = "1860000"+ String.format("%04d",random.nextInt(10000));
            String callIn = "1890000"+ String.format("%04d",random.nextInt(10000));
            String callStatus = arr[random.nextInt(arr.length)];
            int callDuration = "success".equals(callStatus) ? (1 + random.nextInt(10)) * 1000 : 0;

            // 随机产生一条基站日志数据
            StationLog stationLog = new StationLog(
                    "station_" + random.nextInt(10),
                    callOut,
                    callIn,
                    callStatus,
                    System.currentTimeMillis(),
                    callDuration
            );

            // 将一个对象变为json
            String jsonString = JSON.toJSONString(stationLog);
            System.out.println(jsonString);
            ProducerRecord<String, String> producerRecord = new ProducerRecord<String, String>("topicA",jsonString);
            kafkaProducer.send(producerRecord);
            Thread.sleep(2000);


        }


        //kafkaProducer.close();

    }
}

也可以 python 发送 kafka 数据（将以上 java 代码替换为 python 代码）

首先安装 kafka 环境：

pip install kafka-python

接着编写代码：

python 复制代码

from kafka import KafkaProducer
from kafka.errors import KafkaError
import json
import random
import time

class StationLog:
    def __init__(self, station_id, call_out, call_in, call_status, timestamp, call_duration):
        self.station_id = station_id
        self.call_out = call_out
        self.call_in = call_in
        self.call_status = call_status
        self.timestamp = timestamp
        self.call_duration = call_duration

    def to_string(self):
        return json.dumps(self.__dict__)

def main():
    # 设置连接kafka集群的ip和端口
    producer = KafkaProducer(bootstrap_servers='bigdata01:9092',
                             value_serializer=lambda v: json.dumps(v).encode('utf-8'))

    arr = ["fail", "busy", "barring", "success", "success", "success", "success", "success", "success", "success", "success", "success"]

    while True:
        call_out = "1860000" + str(random.randint(0, 9999)).zfill(4)
        call_in = "1890000" + str(random.randint(0, 9999)).zfill(4)
        call_status = random.choice(arr)
        call_duration = 1000 * (10 + random.randint(0, 9)) if call_status == "success" else 0

        # 随机产生一条基站日志数据
        station_log = StationLog(
            "station_" + str(random.randint(0, 9)),
            call_out,
            call_in,
            call_status,
            int(time.time() * 1000),  # 当前时间戳
            call_duration
        )
        print(station_log.to_string())
        time.sleep(0.1 + random.randint(0, 99) / 100)

        try:
            # 发送数据到Kafka
            producer.send('topicA', station_log.to_string())
        except KafkaError as e:
            print(f"Failed to send message: {e}")

        # 确保所有异步消息都被发送
        producer.flush()

if __name__ == "__main__":
    main()

可以使用本地的 kafka-ui 的工具进行消费，查看是否可以正常发送和接收消息：

解压kafka-ui安装包，双击打开bin目录下的kafkaUI.bat（注意：一直保持打开的状态，不要关掉）

通过端口http://localhost:8889/#/进入ui界面

接着编写 pyspark 中的 StructStreaming 代码：

python 复制代码

import os

from pyspark.sql import SparkSession
import pyspark.sql.functions as F

if __name__ == '__main__':
    # 配置环境
    os.environ['JAVA_HOME'] = 'C:/Program Files/Java/jdk1.8.0_241'
    # 配置Hadoop的路径，就是前面解压的那个路径
    os.environ['HADOOP_HOME'] = 'D:/hadoop-3.3.1'
    # 配置base环境Python解析器的路径
    os.environ['PYSPARK_PYTHON'] = 'C:/ProgramData/Miniconda3/python.exe'  # 配置base环境Python解析器的路径
    os.environ['PYSPARK_DRIVER_PYTHON'] = 'C:/ProgramData/Miniconda3/python.exe'

    spark = SparkSession.builder.master("local[2]").appName("streamingkafka").config(
        "spark.sql.shuffle.partitions", 2).getOrCreate()

    readDf = spark.readStream \
       .format("kafka") \
       .option("kafka.bootstrap.servers", "bigdata01:9092") \
       .option("subscribe", "topicA") \
       .load()

    #etlDf = readDf.selectExpr("CAST(value AS STRING)").filter(F.col("value").contains("success"))
    readDf.createOrReplaceTempView("temp_donghu")
    etlDf = spark.sql("""
        select cast(value as string) from temp_donghu where cast(value as string) like '%success%'
    """)


    etlDf.writeStream \
        .format("kafka") \
        .option("kafka.bootstrap.servers", "bigdata01:9092") \
        .option("topic", "etlTopic") \
        .option("checkpointLocation", "./ckp") \
        .start().awaitTermination()

    spark.stop()

cast 函数可以将一个 byte[] 结构的字符串变为一个普通的字符串。

报错 : org.apache.spark.sql.AnalysisException: Failed to find data source: kafka. Please deploy the application as per the deployment section of "Structured Streaming + Kafka Integration Guide".;

解决：这个是因为缺少了Kafka和Spark的集成包，前往https://mvnrepository.com/artifact/org.apache.spark

下载对应的jar包即可，比如我是SparkSql写入的Kafka，那么我就需要下载Spark-Sql-Kafka.x.x.x.jar

接着运行又报错：

Caused by: java.lang.ClassNotFoundException: org.apache.kafka.common.serialization.ByteArraySerializer
	at java.net.URLClassLoader.findClass(URLClassLoader.java:382)
	at java.lang.ClassLoader.loadClass(ClassLoader.java:418)
	at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:355)
	at java.lang.ClassLoader.loadClass(ClassLoader.java:351)
	... 21 more

原因是缺少 kafka-clients.jar 包：

遇到如下错误：

Caused by: java.lang.ClassNotFoundException: org.apache.spark.kafka010.KafkaConfigUpdater
	at java.net.URLClassLoader.findClass(URLClassLoader.java:382)
	at java.lang.ClassLoader.loadClass(ClassLoader.java:418)
	at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:355)
	at java.lang.ClassLoader.loadClass(ClassLoader.java:351)
	... 44 more

在Spark 3.0.0环境下运行StructuredStreaming程序时遇到版本不兼容错误，需要额外添加commons-pools2和spark-token-provider-kafka jar包

将这些 jar 包都下载下来，放入 pyspark 中的 jars 目录下，代码即可运行

情况解决！！！