1、POM文件导入

<dependency>
    <groupId>org.apache.kafka</groupId>
    <artifactId>kafka_2.12</artifactId>
    <version>2.1.1</version>
</dependency>

2、yml配置文件这样写

kafka:
  consumer:
    enable-auto-commit: true
    group-id: kafkaProducer
    auto-commit-interval: 1000
    auto-offset-reset: latest
    bootstrap-servers: ip:prot
    key-deserializer: org.apache.kafka.common.serialization.StringDeserializer
    value-deserializer: org.apache.kafka.common.serialization.StringDeserializer
    concurrency: 3

3、kafka配置类

package com.harzone.kafka;

import org.apache.kafka.common.serialization.StringDeserializer;
import org.springframework.beans.factory.annotation.Value;
import org.apache.kafka.clients.consumer.KafkaConsumer;
import org.springframework.boot.ApplicationArguments;
import org.springframework.boot.ApplicationRunner;
import org.springframework.stereotype.Component;

import java.util.Arrays;
import java.util.Properties;

@Component
public class KafkaConsumerConfig implements ApplicationRunner {
    @Value("${kafka.consumer.bootstrap-servers}")
    private String servers;
    @Value("${kafka.consumer.enable-auto-commit}")
    private boolean enableAutoCommit;
    @Value("${kafka.consumer.auto-commit-interval}")
    private String autoCommitInterval;
    @Value("${kafka.consumer.group-id}")
    private String groupId;
    @Value("${kafka.consumer.auto-offset-reset}")
    private String autoOffsetReset;
    @Value("${kafka.consumer.concurrency}")
    private int concurrency;
    private final String topic = "test_producer";
    Properties props;
    private KafkaConsumer<String, String> kafkaConsumer;

    private Properties initProperties() {
        // zookeeper 配置
        props = new Properties();
        props.put("bootstrap.servers", servers);
        // group 代表一个消费组
        props.put("group.id", "kafkaProducer");
        props.put("session.timeout.ms", "30000");
        // 往zookeeper上写offset的频率
        props.put("auto.commit.interval.ms", "1000");
        // key的反序列化类型
        props.put("key.deserializer", StringDeserializer.class.getName());
        props.put("value.deserializer", StringDeserializer.class.getName());
        return props;

    }

    @Override
    public void run(ApplicationArguments args) throws Exception {
        //初始化kafka链接
        initProperties();
        //Runtime.getRuntime().availableProcessors()线程数量
        KafkaConsumerThread consumerThread = new KafkaConsumerThread(props, topic, Runtime.getRuntime().availableProcessors() - 1);
        consumerThread.start();
    }
}
4、kafka多线程拉取实现

package com.harzone.kafka;

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;

import java.time.Duration;
import java.util.*;
import java.util.concurrent.ArrayBlockingQueue;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.ThreadPoolExecutor;
import java.util.concurrent.TimeUnit;

public class KafkaConsumerThread extends Thread {

    private static final Logger log = LoggerFactory.getLogger(KafkaConsumerThread.class);

    private KafkaConsumer<String, String> kafkaConsumer;
    //线程池
    private ExecutorService executorService;
    //线程数
    private int threadNumber;

    static List<String> list = new ArrayList<>();

    //有参构造函数，初始化数据配置
    public KafkaConsumerThread(Properties properties, String topic, int availableProcessors) {
        kafkaConsumer = new KafkaConsumer<String, String>(properties);
        //subscribe从最新处消费（assign从最后处消费）----singletonList:返回一个不可变的列表
        kafkaConsumer.subscribe(Collections.singletonList(topic));
        this.threadNumber = availableProcessors;
        //ThreadPoolExecutor线程池（核心线程池大小，最大线程数，线程最大空闲时间，时间单位，线程等待队列，拒绝策略）
        executorService = new ThreadPoolExecutor(threadNumber, threadNumber, 0L, TimeUnit.MILLISECONDS,
                new ArrayBlockingQueue<>(1000), new ThreadPoolExecutor.CallerRunsPolicy());
    }

    @Override
    public void run() {
        try {
            while (true) {
                //poll(多长时间拉取一次ms)
                ConsumerRecords<String, String> records = kafkaConsumer.poll(Duration.ofMillis(10));
                if (!records.isEmpty()) {
                    executorService.submit(new RecordsHandler(records));
                }
            }
        } catch (Exception e) {
            e.printStackTrace();
        } finally {
            kafkaConsumer.close();
        }
    }
}