RocketMQ Connect实战3

RocketMQ Connect Overview

准备

启动RocketMQ

  1. Linux/Unix/Mac
  2. 64bit JDK 1.8+;
  3. Maven 3.2.x或以上版本;
  4. 启动 RocketMQ;

tips : ${ROCKETMQ_HOME} 位置说明

bin-release.zip 版本：/rocketmq-all-4.9.4-bin-release

source-release.zip 版本：/rocketmq-all-4.9.4-source-release/distribution

启动Connect

Connector插件编译

Debezium RocketMQ Connector

  1. $ cd rocketmq-connect/connectors/rocketmq-connect-debezium/
  2. $ mvn clean package -Dmaven.test.skip=true

将 Debezium MySQL PostgreSQL RocketMQ Connector 编译好的包放入Runtime加载目录。命令如下：

  1. mkdir -p /usr/local/connector-plugins
  2. cp rocketmq-connect-debezium-postgresql/target/rocketmq-connect-debezium-postgresql-0.0.1-SNAPSHOT-jar-with-dependencies.jar /usr/local/connector-plugins
  4. cp rocketmq-connect-debezium-mysql/target/rocketmq-connect-debezium-mysql-0.0.1-SNAPSHOT-jar-with-dependencies.jar /usr/local/connector-plugins

JDBC Connector

将 JDBC Connector 编译好的包放入Runtime加载目录。命令如下：

  1. $ cd rocketmq-connect/connectors/rocketmq-connect-jdbc/
  2. $ mvn clean package -Dmaven.test.skip=true
  3. cp rocketmq-connect-jdbc/target/rocketmq-connect-jdbc-0.0.1-SNAPSHOT-jar-with-dependencies.jar /usr/local/connector-plugins

启动Connect Runtime

  1. cd  rocketmq-connect
  3. mvn -Prelease-connect -DskipTests clean install -U

修改配置connect-standalone.conf ，重点配置如下

  1. $ cd distribution/target/rocketmq-connect-0.0.1-SNAPSHOT/rocketmq-connect-0.0.1-SNAPSHOT
  2. $ vim conf/connect-standalone.conf
  1. workerId=standalone-worker
  2. storePathRootDir=/tmp/storeRoot
  4. ## Http port for user to access REST API
  5. httpPort=8082
  7. # Rocketmq namesrvAddr
  8. namesrvAddr=localhost:9876
  10. # RocketMQ acl
  11. aclEnable=false
  12. accessKey=rocketmq
  13. secretKey=12345678
  15. autoCreateGroupEnable=false
  16. clusterName="DefaultCluster"
  18. # 核心配置，将之前编译好debezium包的插件目录配置在此；
  19. # Source or sink connector jar file dir,The default value is rocketmq-connect-sample
  20. pluginPaths=/usr/local/connector-plugins
  1. cd distribution/target/rocketmq-connect-0.0.1-SNAPSHOT/rocketmq-connect-0.0.1-SNAPSHOT
  3. sh bin/connect-standalone.sh -c conf/connect-standalone.conf &

Postgres镜像

使用debezium的Postgres docker搭建环境MySQL数据库

  1. # starting a pg instance
  2. docker run -d --name postgres -p 5432:5432 -e POSTGRES_USER=start_data_engineer -e POSTGRES_PASSWORD=password debezium/postgres:14
  4. # bash into postgres instance
  5. docker exec -ti postgres /bin/bash

Postgres信息 端口：5432 账号：start_data_engineer/password 同步的源数据库：bank.user

MySQL镜像

使用debezium的MySQL docker搭建环境MySQL数据库

  1. docker run -it --rm --name mysql -p 3306:3306 -e MYSQL_ROOT_PASSWORD=debezium -e MYSQL_USER=mysqluser -e MYSQL_PASSWORD=mysqlpw quay.io/debezium/example-mysql:1.9

MySQL信息

端口：3306

账号：root/debezium 同步的源数据库：bank.user

目标库：bank1.user

测试数据

通过root/debezium账号登录数据库 源数据库表：bank.user

  1. create database bank;
  2. use bank;
  4. create table bank.user
  5. (
  6.     id       bigint NOT NULL AUTO_INCREMENT,
  7.     user_id          integer,
  8.     name    varchar(8),
  9.     age integer,
  10.     birthday date,
  11.     datetime_created timestamp(3),
  12.     datetime_updated timestamp(3),
  13.     height           decimal(11, 2) null,
  14.     PRIMARY KEY (`id`)
  15. );
  17. insert into bank.user values (1003, 1, 'lilei2', 10, now(), now(), now(), 1.72);
  18. update bank.user set user_id = 1003 where id = 1003;

通过start_data_engineer/password账号登录PostgreSQL数据库

源数据库表：bank.user

  1. CREATE SCHEMA bank;
  2. SET search_path TO bank,public;
  3. create table bank.user
  4. (
  5.     id       integer not null
  6.         constraint user_pkey
  7.             primary key,
  8.     user_id          integer,
  9.     name    varchar(8),
  10.     age integer,
  11.     birthday date,
  12.     datetime_created timestamp(3),
  13.     datetime_updated timestamp(3),
  14.     height           numeric(11, 2)
  15. );
  17. insert into bank.user values (1001, 1, 'lilei1', 10, now(), now(), now(), 1.72);
  18. update bank.user set user_id = 1001 where id = 1001;

目标表：bank1.user

  1. create database bank1;
  2. create table bank1.user
  3. (
  4.     id               bigint auto_increment
  5.         primary key,
  6.     user_id          int            null,
  7.     name             varchar(8)     null,
  8.     age              int            null,
  9.     birthday         date           null,
  10.     datetime_created timestamp(3)   null,
  11.     datetime_updated timestamp(3)   null,
  12.     height           decimal(11, 2) null
  13. );

启动Connector

启动Debezium source connector

同步原表数据：bank.user 作用：通过解析MySQL binlog 封装成通用的ConnectRecord对象，发送的RocketMQ Topic当中

  1. curl -X POST -H "Content-Type: application/json" http://127.0.0.1:8082/connectors/MySQLCDCSource1000 -d '{
  2. "connector.class": "org.apache.rocketmq.connect.debezium.mysql.DebeziumMysqlConnector",
  3. "max.task": "1",
  4. "connect.topicname": "debezium-source-topic1000",
  5. "kafka.transforms": "Unwrap",
  6. "kafka.transforms.Unwrap.delete.handling.mode": "none",
  7. "kafka.transforms.Unwrap.type": "io.debezium.transforms.ExtractNewRecordState",
  8. "kafka.transforms.Unwrap.add.headers": "op,source.db,source.table",
  9. "database.history.skip.unparseable.ddl": true,
  10. "database.history.name.srv.addr": "localhost:9876",
  11. "database.history.rocketmq.topic": "db-history-debezium-topic1000",
  12. "database.history.store.only.monitored.tables.ddl": true,
  13. "include.schema.changes": false,
  14. "database.server.name": "dbserver1",
  15. "database.port": 3306,
  16. "database.hostname": "数据库ip",
  17. "database.connectionTimeZone": "UTC",
  18. "database.user": "debezium",
  19. "database.password": "dbz",
  20. "table.include.list": "bank.user",
  21. "max.batch.size": 50,
  22. "database.include.list": "bank",
  23. "snapshot.mode": "when_needed",
  24. "database.server.id": "184054",
  25. "key.converter": "org.apache.rocketmq.connect.runtime.converter.record.json.JsonConverter",
  26. "value.converter": "org.apache.rocketmq.connect.runtime.converter.record.json.JsonConverter"
  27. }'

同步原表数据：bank.user 作用：通过解析Postgres binlog 封装成通用的ConnectRecord对象，发送的RocketMQ Topic当中

  1. curl -X POST -H "Content-Type: application/json" http://127.0.0.1:8082/connectors/postgres-connector1000 -d  '{
  2.   "connector.class": "org.apache.rocketmq.connect.debezium.postgres.DebeziumPostgresConnector",
  3.   "max.task": "1",
  4.   "connect.topicname": "debezium-source-topic1000",
  5.   "kafka.transforms": "Unwrap",
  6.   "kafka.transforms.Unwrap.delete.handling.mode": "none",
  7.   "kafka.transforms.Unwrap.type": "io.debezium.transforms.ExtractNewRecordState",
  8.   "kafka.transforms.Unwrap.add.headers": "op,source.db,source.table",
  9.   "database.history.skip.unparseable.ddl": true,
  10.   "database.server.name": "bankserver1",
  11.   "database.port": 5432,
  12.   "database.hostname": "数据库ip",
  13.   "database.connectionTimeZone": "UTC",
  14.   "database.user": "start_data_engineer",
  15.   "database.dbname": "start_data_engineer",
  16.   "database.password": "password",
  17.   "table.whitelist": "bank.user",
  18.   "key.converter": "org.apache.rocketmq.connect.runtime.converter.record.json.JsonConverter",
  19.   "value.converter": "org.apache.rocketmq.connect.runtime.converter.record.json.JsonConverter"
  20. }'

启动 jdbc sink connector

作用：通过消费Topic中的数据，通过JDBC协议写入到目标表当中

  1. curl -X POST -H "Content-Type: application/json" http://127.0.0.1:8082/connectors/jdbcmysqlsinktest1000 -d '{
  2.   "connector.class": "org.apache.rocketmq.connect.jdbc.connector.JdbcSinkConnector",
  3.   "max.task": "2",
  4.   "connect.topicnames": "debezium-source-topic1000",
  5.   "connection.url": "jdbc:mysql://数据库ip:3306/bank1",
  6.   "connection.user": "root",
  7.   "connection.password": "debezium",
  8.   "pk.fields": "id",
  9.   "table.name.from.header": "true",
  10.   "pk.mode": "record_key",
  11.   "insert.mode": "UPSERT",
  12.   "db.timezone": "UTC",
  13.   "table.types": "TABLE",
  14.   "errors.deadletterqueue.topic.name": "dlq-topic",
  15.   "errors.log.enable": "true",
  16.   "errors.tolerance": "ALL",
  17.   "delete.enabled": "true",
  18.   "key.converter": "org.apache.rocketmq.connect.runtime.converter.record.json.JsonConverter",
  19.   "value.converter": "org.apache.rocketmq.connect.runtime.converter.record.json.JsonConverter"
  20. }'

以上三个Connector任务创建成功以后 通过start_data_engineer/password账号登录PostgreSQL数据库 账号登录数据库 或者通过root/debezium账号登录MySQL数据库

对源数据库表：bank.user增删改 都会同步到同步到目标表MySQL bank1.user