debezium、kafka connector 解析 mysql binlog 到 kafak

目的： 需要搭建一个可以自动监听MySQL数据库的变化，将变化的数据捕获处理，此处只讲解如何自动捕获mysql 中数据的变化

使用的技术

debezium ：https://debezium.io/documentation/reference/1.0/connectors/mysql.html

kafka: http://kafka.apache.org/

zookeeper :  http://zookeeper.apache.org/

mysql 5.7  https://www.mysql.com/

一、思路

需要一台 Centos 7.x 的虚拟机 ，zk、debezium、kafka、confluent 运行在 虚拟机上 ，mysql 运行在 windows 系统上，虚拟机监听 window 环境下的 mysql 数据变化

二、MySQL 环境准备

首先需要找到 mysql 的配置文件：my.ini ,我的路径是：C:ProgramDataMySQLMySQL Server 5.7 ，因为监听基础是基于 mysql binlog ，需要开启binlog ，添加如下配置

log_bin =D:mysql-binlogmysql-bin

binlog_format=Row

server-id=223344

binlog_row_image  = full

expire_logs_days  = 10

binlog_rows_query_log_events = on

重启 mysql 服务

net stop  mysql57
net start  mysql57

此处，MySQL binlog 即开启，可以简单的验证,cmd 窗口 mysql -u root -p 登录 mysql 

show binary logs;

 可以看到文件内容，即mysql 变化的二进制文件。到此处，MySQL准备就绪。

二、zookeeper 、 kafka  准备

下载 zookeeper-3.4.14.tar.gz 、kafka_2.12-2.2.0.tar

mkdir -p  /usr/local/software/zookeeper
mkdir -p  /usr/local/software/kafka
mkdir -p  /usr/local/software/confluent

准备好路径,并将安装包移入该目录，并解压

mv  zookeeper-3.4.14.tar.gz   /usr/local/software/zookeeper
mv  kafka_2.12-2.2.0.tar

进入 zookeeper   /usr/local/software/zookeeper/zookeeper-3.4.14/conf目录，修改 zoo.cfg （原名 zoo_sample.cfg)内容

dataDir=/opt/data/zookeeper/data
dataLogDir=/opt/data/zookeeper/logs

进入 dataDir 目录，创建文件 myid ,并添加内容:  1

此处，zk 的配置修改结束。开启配置 kafka  路径是：/usr/local/software/kafka/kafka_2.12-2.2.0/config， 修改 server.properties 

broker.id=1
listeners=PLAINTEXT://192.168.91.25:9092
advertised.listeners=PLAINTEXT://192.168.91.25:9092
log.dirs=/opt/data/kafka-logs
host.name=192.168.91.25
zookeeper.connect=localhost:2181

三、debezium配置

此处需要 debezium connector 对 mysql 的 jar 包，下载地址：https://debezium.io/releases/1.0/

 将下载好的 plugs 上传到虚拟机，解压后名称是： debezium-connector-mysql

移动到： /usr/local/share/kafka/plugins 目录下，如果没有该目录则手动创建

 依赖的 jar 包下载好后，配置 kafka 目录中conf connector

目录： /usr/local/software/kafka/kafka_2.12-2.2.0/conf/connect-standalone.properties

bootstrap.servers=本机IP:9092
plugin.path=/usr/local/share/kafka/plugins

 此外，在kafka 根目录下 创建文件： msyql.properties ,内容

name=mysql
connector.class=io.debezium.connector.mysql.MySqlConnector
database.hostname=192.168.3.125
database.port=3306
database.user=root
database.password=123456
database.server.id=112233
database.server.name=test  
database.whitelist=orders,users  
database.history.kafka.bootstrap.servers=192.168.91.25:9092
database.history.kafka.topic=history.test
include.schema.changes=true
include.query=true
# options: adaptive_time_microseconds（default）adaptive(deprecated) connect()
time.precision.mode=connect
# options: precise(default) double string
decimal.handling.mode=string
# options: long(default) precise
bigint.unsigned.handling.mode=long

四、启动服务

01.启动zk

cd /usr/local/software/zookeeper/zookeeper-3.4.14 

zkServer.sh  start

02.启动kafka

cd  /usr/local/software/kafka/kafka_2.12-2.2.0

./bin/kafka-server-start.sh  -daemon  config/server.properties 

03.启动kafka  connector

cd  /usr/local/software/kafka/kafka_2.12-2.2.0

 ./bin/connect-standalone.sh  config/connect-standalone.properties  mysql.properties 

04.查看 topic ,在新的端口查看

 ./bin/kafka-topics.sh --list --zookeeper localhost:2181

五、指定监听的数据库/表

在 postman 中模拟 post 请求，以 json 格式传递参数：表示 监听 shiro数据库

{
  "name": "shiro",  
  "config": {
    "connector.class": "io.debezium.connector.mysql.MySqlConnector", 
    "database.hostname": "192.168.3.125", 
    "database.port": "3306", 
    "database.user": "root", 
    "database.password": "123456", 
    "database.server.id": "184054", 
    "database.server.name": "my", 
    "database.whitelist": "shiro", 
    "database.history.kafka.bootstrap.servers": "192.168.91.25:9092", 
    "database.history.kafka.topic": "history.shiro", 
    "include.schema.changes": "true" 
  }}

重新查看topic 

在新端口启动 kafka 消费者，消费my.shiro.user 

./bin/kafka-console-consumer.sh  --bootstrap-server localhost:9092 --topic my.shiro.user --from-beginning

Java客户端消费者代码

package kafka;

import org.apache.kafka.clients.consumer.ConsumerConfig;
import org.apache.kafka.clients.consumer.ConsumerRecord;
import org.apache.kafka.clients.consumer.ConsumerRecords;
import org.apache.kafka.clients.consumer.KafkaConsumer;

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

/**
 * Created by baizhuang on 2019/10/25 10:39.
 */

public class MyConsumer {
    public static void main(String []args){

        //1.创建 kafka 生产者配置信息。
        Properties properties = new Properties();

        //2.指定 kafka 集群
        properties.put(ConsumerConfig.BOOTSTRAP_SERVERS_CONFIG,"192.168.91.25:9092");

        properties.put(ConsumerConfig.ENABLE_AUTO_COMMIT_CONFIG,true);

        properties.put(ConsumerConfig.AUTO_COMMIT_INTERVAL_MS_CONFIG,"1000");

        //key,value 反序列化
        properties.put(ConsumerConfig.KEY_DESERIALIZER_CLASS_CONFIG,"org.apache.kafka.common.serialization.StringDeserializer");

        properties.put(ConsumerConfig.VALUE_DESERIALIZER_CLASS_CONFIG,"org.apache.kafka.common.serialization.StringDeserializer");

        properties.put("group.id","test");


        KafkaConsumer<String,String> consumer = new  KafkaConsumer<String,String>(properties);


        consumer.subscribe(Arrays.asList("my.shiro.user"));


        while (true) {
            ConsumerRecords<String, String> consumerRecords = consumer.poll(100);

            for (ConsumerRecord<String, String> consumerRecord : consumerRecords) {
                System.out.println(consumerRecord.key() + "-----" + consumerRecord.value());
            }

        }

    }
}

Java 客户端生产者代码

package kafka;

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;

/**
 * Created by baizhuang on 2019/10/24 16:58.
 */

public class MyProducer {
    public static void main(String []args){


        //1.创建 kafka 生产者配置信息。
        Properties properties = new Properties();

        //2.指定 kafka 集群
        properties.put(ProducerConfig.BOOTSTRAP_SERVERS_CONFIG,"192.168.91.25:9092");

        //3.
        properties.put("acks","all");

        //4.重试次数
        properties.put("retries",0);

        //5.批次大小
        properties.put("batch.size",16384);

        //6.等待时间
        properties.put("linger.ms",1);

        //7.RecordAccumlate 缓冲区大小
        properties.put("buffer.memory",33554432);

        //key ,value 序列化
        properties.put("key.serializer", "org.apache.kafka.common.serialization.StringSerializer");
        properties.put("value.serializer", "org.apache.kafka.common.serialization.StringSerializer");


        //9.创建生产者
        KafkaProducer<String,String>  producer = new KafkaProducer<String, String>(properties);

        for(int i=0;i<10;i++){
            //10.发送
            String key = String.valueOf(i);
            String value = "第"+key+"条消息";
            producer.send(new ProducerRecord<String, String>("mytopic",key,value));
            System.out.println("msg:"+i);
        }

        producer.close();

    }
}

启动消费者，修改 shiro 数据库的user 表，Java客户端消费者与 linux 消费者均可动态的显示变化的数据