Question

设置：：我在MS SQL Server上启用了CDC，并且使用debezium kafka connect（source）将CDC事件馈送到Kafka。同样，多个表CDC事件也被路由到Kafka中的单个主题。

问题：由于我在kafka主题中拥有多个表数据，因此我希望CDC数据中包含表名和数据库名。

我在MySQL CDC中获得表名和数据库名，而在MS SQL CDC中却没有。

下面是SQL Server的Debezium源连接器

curl -i -X POST -H "Accept:application/json" -H "Content-Type:application/json" localhost:8083/connectors/ -d '{
  "name": "cdc-user_profile-connector",
  "config": {
    "connector.class": "io.debezium.connector.sqlserver.SqlServerConnector",
    "tasks.max": "1",
    "database.hostname": "<<hostname>>",
    "database.port": "<<port>>",
    "database.user": "<<username>>",
    "database.password": "<<password>>",
    "database.server.name": "test",
    "database.dbname": "testDb",
    "table.whitelist": "schema01.table1,schema01.table2",
    "database.history.kafka.bootstrap.servers": "broker:9092",
    "database.history.kafka.topic": "digital.user_profile.schema.audit",
    "database.history.store.only.monitored.tables.ddl": true,
    "include.schema.changes": false,
    "event.deserialization.failure.handling.mode": "fail",
    "snapshot.mode": "initial_schema_only",
    "snapshot.locking.mode": "none",
    "transforms":"addStaticField,topicRoute",
    "transforms.addStaticField.type":"org.apache.kafka.connect.transforms.InsertField$Value",
    "transforms.addStaticField.static.field":"source_system",
    "transforms.addStaticField.static.value":"source_system_1",
    "transforms.topicRoute.type":"org.apache.kafka.connect.transforms.RegexRouter",
    "transforms.topicRoute.regex":"(.*)",
    "transforms.topicRoute.replacement":"digital.user_profile",
    "errors.tolerance": "none",
    "errors.log.enable": true,
    "errors.log.include.messages": true,
    "errors.retry.delay.max.ms": 60000,
    "errors.retry.timeout": 300000
  }
}'

我得到以下输出（演示数据）

{
  "before": {
    "profile_id": 147,
    "email_address": "test@gmail.com"
  },
  "after": {
    "profile_id": 147,
    "email_address": "test_modified@gmail.com"
  },
  "source": {
    "version": "0.9.4.Final",
    "connector": "sqlserver",
    "name": "test",
    "ts_ms": 1556723528917,
    "change_lsn": "0007cbe5:0000b98c:0002",
    "commit_lsn": "0007cbe5:0000b98c:0003",
    "snapshot": false
  },
  "op": "u",
  "ts_ms": 1556748731417,
  "source_system": "source_system_1"
}

我的要求是达到以下目标

{
  "before": {
    "profile_id": 147,
    "email_address": "test@gmail.com"
  },
  "after": {
    "profile_id": 147,
    "email_address": "test_modified@gmail.com"
  },
  "source": {
    "version": "0.9.4.Final",
    "connector": "sqlserver",
    "name": "test",
    "ts_ms": 1556723528917,
    "change_lsn": "0007cbe5:0000b98c:0002",
    "commit_lsn": "0007cbe5:0000b98c:0003",
    "snapshot": false,
    "db": "testDb",
    "table": "table1/table2"
  },
  "op": "u",
  "ts_ms": 1556748731417,
  "source_system": "source_system_1"
}

Answer 1

Debezium Kafka-Connect通常将每个表中的数据放在单独的主题中，主题名称的格式为hostname.database.table。通常，我们使用主题名称来区分源表和数据库名称。

如果要将所有表中的数据手动放入一个主题，则可能还必须手动添加表和数据库名称。

Answer 2

这计划是https://issues.jboss.org/browse/DBZ-875问题的一部分

如何从debezium kafka connect收到的CDC事件中获取表名和数据库名

2 个答案: