序言

本文使用的是debezium连接器来实时同步SQL server数据，再写入到kafka消息队列中。

一、sqlserver开启CDC机制

1、开启 SQL Server Agent 服务，确保SQL server 代理 服务正常运行
2、开启cdc机制

-- 执行下面sql语句

-- 创建表
CREATE TABLE [dbo].[student] (
  [name] char(25) COLLATE Chinese_PRC_CI_AS  NOT NULL,
  [sex] char(10) COLLATE Chinese_PRC_CI_AS  NOT NULL,
  [id] int  NOT NULL,
  [age] int  NOT NULL
)
GO

-- 开启cdc
EXEC sys.sp_cdc_enable_db 
GO

EXEC sys.sp_cdc_enable_table
    @source_schema = N'dbo'
  , @source_name = N'student' -- 表名
  , @role_name = N'cdc_admin'--增加的角色
  , @capture_instance = N'student_instance'--实例名 
  , @supports_net_changes = 1
  , @filegroup_name = N'PRIMARY';
GO

-- 查看数据库是否启用cdc
SELECT name,is_cdc_enabled FROM sys.databases WHERE is_cdc_enabled = 1;

-- 查看表是否启用cdc
SELECT name,is_tracked_by_cdc FROM sys.tables WHERE is_tracked_by_cdc = 1;

-- 有结果则说明开启成功

二、配置connector

1、下载 debezium-connector-sqlserver 下载地址

2、配置 $KAFKA_HOME/config/connect-distributed.properties，详见实时同步Oracle数据，写入Kafka

3、解压下载的tar包到 connectors目录中

三、启动connector

cd /usr/software/kafka/kafka_2.12-2.4.0/config/

connect-distributed.sh ./connect-distributed.properties

# 确认是否正确配置connector
curl -s node1:18083/connector-plugins | jq

# 有下面结果说明成功
{
  "class": "io.debezium.connector.sqlserver.SqlServerConnector",
  "type": "source",
  "version": "1.4.1.Final"
},

# 启动connector的两种方法
# 方法1
curl -s -X POST -H "Content-Type: application/json" --data '
{
 "name": "sqlserver-kafka-connector",
 "config": {
     "connector.class" : "io.debezium.connector.sqlserver.SqlServerConnector",
     "tasks.max" : "1",
     "database.server.name" : "server1",
     "database.hostname" : "hostname",
     "database.port" : "1433",
     "database.user" : "username",
     "database.password" : "password",
     "database.dbname" : "CDCTest",
     "database.history.kafka.bootstrap.servers" : "node1:9092,node2:9092,node3:9092",
     "database.history.kafka.topic": "schema-changes.inventory"
   }
}' http://node1:18083/connectors | jq

# 方法2
curl -i -X POST -H "Accept:application/json" -H "Content-Type:application/json" http://node1:18083/connectors/ -d @register-sqlserver.json

# register-sqlserver.json 内容如下
{
 "name": "sqlserver-kafka-connector",
 "config": {
     "connector.class" : "io.debezium.connector.sqlserver.SqlServerConnector",
     "tasks.max" : "1",
     "database.server.name" : "server1",
     "database.hostname" : "hostname",
     "database.port" : "1433",
     "database.user" : "username",
     "database.password" : "password",
     "database.dbname" : "CDCTest",
     "database.history.kafka.bootstrap.servers" : "node1:9092,node2:9092,node3:9092",
     "database.history.kafka.topic": "schema-changes.inventory"
   }
}
注意:方法2 需要进入到存放 register-sqlserver.json 的目录中

# 启动成功后可以查看到一下内容，说明启动成功
[root@node1 kafka]# curl -s node1:18083/connectors | jq
[
  "sqlserver-kafka-connector",
  "oracle-kafka-connector"
]
[root@node1 kafka]# curl -s node1:18083/connectors/sqlserver-kafka-connector/status | jq
{
  "name": "sqlserver-kafka-connector",
  "connector": {
    "state": "RUNNING",
    "worker_id": "192.168.2.170:18083"
  },
  "tasks": [
    {
      "id": 0,
      "state": "RUNNING",
      "worker_id": "192.168.2.170:18083"
    }
  ],
  "type": "source"
}

四、同步sqlserver数据到kafka

-- 往student表插入数据
insert into student values(1,'jack',22,'male')

-- server1.dbo.student 为我们需要消费的DML数据
[root@node2 ~]# kafka-topics.sh --list --bootstrap-server node1:9092,node2:9092,node3:9092
TEST.KAFKA.STUDENT
TEST.KAFKA._GENERIC_DDL
__consumer_offsets
connect-configs
connect-offsets
connect-status
schema-changes.inventory
server1
server1.dbo.student
server1.dbo.teacher
shcema-changes.inventory

-- 消费server1.dbo.student
kafka-console-consumer.sh --topic server1.dbo.student --bootstrap-server node1:9092,node2:9092,node3:9092 --from-beginning | jq
{
  "before": null,
  "after": {
    "id": 1,
    "name": "jack                     ",
    "age": 22,
    "sex": "male      "
  },
  "source": {
    "version": "1.4.1.Final",
    "connector": "sqlserver",
    "name": "server1",
    "ts_ms": 1612779384700,
    "snapshot": "false",
    "db": "CDCTest",
    "schema": "dbo",
    "table": "student",
    "change_lsn": "00000030:000002b8:0002",
    "commit_lsn": "00000030:000002b8:0003",
    "event_serial_no": 1
  },
  "op": "c",
  "ts_ms": 1612779385938,
  "transaction": null
}

如有问题，欢迎一起交流讨论。

上一篇:实时同步Oracle数据，写入Kafka