尝试在windows 平台的搭建kafka平台的记录

本文主要操作参考：https://blog.csdn.net/woshixiazaizhe/article/details/80610432?spm=1001.2014.3001.5506

1. kafka内置zookeeper启动报错:ZooKeeper audit is disabled

修改kafka安装目录下config中的ZooKeeper.properties文件，修改该选项：audit.enable=true（如果没有改语句则进行添加）

2. 启动zookeeper后再启动kafka，成功；但是显示有个ERROR

这样进行修改

3.按照所有步骤操作并创建Topic后，在本地继续用python尝试生产和订阅数据

3.1. python生产消息到kafka：

from kafka import KafkaProducer
from kafka.errors import kafka_errors
import traceback
import json

def kafka_producer():
    producer = KafkaProducer(
        bootstrap_servers=['localhost:9092'],  # kafka服务器ip和端口号
        key_serializer=lambda k: json.dumps(k).encode(),  # 假设生产的消息为json字符串
        value_serializer=lambda v: json.dumps(v).encode())  # 假设生产的消息为json字符串

    data = {'msg': 'hello kafka,17:56!'}

    future = producer.send(
        'test',  # 要发送的kafka主题
        key='1',  # 同一个key值，会被送至同一个分区
        value=data,
        partition=0)  # 向分区1发送消息
    print("send {}".format(str(data)))

    try:
        future.get(timeout=10)  # 监控是否发送成功
    except kafka_errors:  # 发送失败抛出kafka_errors
        traceback.format_exc()

if __name__ == '__main__':
    kafka_producer()

3.2. python订阅消息：

import time
from kafka import KafkaConsumer


# Set up Kafka consumer
consumer = KafkaConsumer(
    # 'test',  # Topic to consume from
    'mbt_intelligent_control',
    bootstrap_servers=['localhost:9092'],  # Kafka broker addresses
    auto_offset_reset='earliest',  # Start consuming from the beginning of the topic
    enable_auto_commit=True,  # Commit offsets automatically after consuming messages
    group_id='carbon-taos-consumer'  # Consumer group ID
)

# Consume messages from Kafka topic
# print(message.value.decode('utf-8'))
for message in consumer:
    print(message.value.decode('utf-8'))

3.3. 尝试用go订阅消息：

在cursor的帮助下，成功消费到kafka的消息哈哈哈哈