前提：1、安装docker for Windows，2，安装docker-compose

• 镜像选择：

对比网上给出的版本比较，选择wurstmeister/kafka 和 wurstmeister/zookeeper，

原因：kafka版本较新，zookeeper跟kafka分在2个镜像中

docker-compose.yml

version: '3.7'

services:
  zoo1:
    image: wurstmeister/zookeeper
    restart: unless-stopped
    hostname: zoo1
    ports:
      - "2181:2181"
    container_name: zookeeper

  # kafka version: 1.1.0
  # scala version: 2.12
  kafka1:
    image: wurstmeister/kafka
    ports:
      - "9092:9092"
    environment:
      KAFKA_ADVERTISED_HOST_NAME: localhost
      KAFKA_ZOOKEEPER_CONNECT: "zoo1:2181"
      KAFKA_BROKER_ID: 1
      KAFKA_OFFSETS_TOPIC_REPLICATION_FACTOR: 1
      KAFKA_CREATE_TOPICS: "stream-in:1:1,stream-out:1:1"
    depends_on:
      - zoo1
    container_name: kafka

• 启动容器

docker-compose up -d

• 停止容器

docker-compose down

• 消息发送测试

进入容器
docker exec -it kafka /bin/bash
cd /opt/kafka/bin
创建topic，partition=1，topic为:kafkaTopicDev
kafka-topics.sh --create --zookeeper zoo1:2181 --replication-factor 1 --partitions 1 --topic kafkaTopicDev
查看topic
kafka-topics.sh --zookeeper zoo1:2181 --list
创建一个生产者
kafka-console-producer.sh --broker-list localhost:9092 --topic kafkaTopicDev
创建一个消费者
kafka-console-consumer.sh --bootstrap-server localhost:9092 --topic kafkaTopicDev --from-beginning

生产者发送消息，可以看到消费者能实时接收到消息