实时计算之FlinkCDC

什么是CDC

CDC是Change Data Capture(变更数据获取)的简称。核心思想是&＃xff0c;监测并捕获数据库的变动&＃xff08;包括数据或数据表的插入、更新以及删除等&＃xff09;&＃xff0c;将这些变更按发生的顺序完整记录下来&＃xff0c;写入到消息中间件中以供其他服务进行订阅及消费。

CDC的种类

CDC主要分为基于查询和基于Binlog两种方式&＃xff0c;我们主要了解一下这两种之间的区别&＃xff1a;

Flink-CDC

Flink社区开发了 flink-cdc-connectors 组件&＃xff0c;这是一个可以直接从 MySQL、PostgreSQL 等数据库直接读取全量数据和增量变更数据的 source 组件。目前也已开源&＃xff0c;开源地址&＃xff1a;https://github.com/ververica/flink-cdc-connectors

主要代码

//2.Flink-CDC将读取binlog的位置信息以状态的方式保存在CK,如果想要做到断点续传,需要从Checkpoint或者Savepoint启动程序//2.1 开启Checkpoint,每隔5秒钟做一次CKenv.enableCheckpointing(5000L);//2.2 指定CK的一致性语义env.getCheckpointConfig().setCheckpointingMode(CheckpointingMode.EXACTLY_ONCE);//2.3 设置任务关闭的时候保留最后一次CK数据env.getCheckpointConfig().enableExternalizedCheckpoints(CheckpointConfig.ExternalizedCheckpointCleanup.RETAIN_ON_CANCELLATION);//2.4 指定从CK自动重启策略env.setRestartStrategy(RestartStrategies.fixedDelayRestart(3, 2000L));//2.5 设置状态后端env.setStateBackend(new FsStateBackend("hdfs://hadoop102:8020/flinkCDC"));//2.6 设置访问HDFS的用户名System.setProperty("HADOOP_USER_NAME", "bigdata");//3.创建Flink-MySQL-CDC的SourceProperties properties &＃61; new Properties();//initial (default): Performs an initial snapshot on the monitored database tables upon first startup, and continue to read the latest binlog.//latest-offset: Never to perform snapshot on the monitored database tables upon first startup, just read from the end of the binlog which means only have the changes since the connector was started.//timestamp: Never to perform snapshot on the monitored database tables upon first startup, and directly read binlog from the specified timestamp. The consumer will traverse the binlog from the beginning and ignore change events whose timestamp is smaller than the specified timestamp.//specific-offset: Never to perform snapshot on the monitored database tables upon first startup, and directly read binlog from the specified offset.properties.setProperty("scan.startup.mode", "initial");DebeziumSourceFunction mysqlSource &＃61; MySQLSource.builder().hostname("hadoop102").port(3306).username("root").password("000000").databaseList("test").tableList("test.z_user_info") //可选配置项,如果不指定该参数,则会读取上一个配置下的所有表的数据
//注意&＃xff1a;指定的时候需要使用"db.table"的方式.debeziumProperties(properties).deserializer(new StringDebeziumDeserializationSchema()).build();

给当前的Flink程序创建Savepoint

[bigdata&＃64;hadoop102 flink-standalone]$ bin/flink savepoint JobId hdfs://hadoop102:8020/flink/save


FlinkSQL方式的应用

//2.创建Flink-MySQL-CDC的SourcetableEnv.executeSql("CREATE TABLE user_info (" &＃43;" id INT," &＃43;" name STRING," &＃43;" phone_num STRING" &＃43;") WITH (" &＃43;" &＃39;connector&＃39; &＃61; &＃39;mysql-cdc&＃39;," &＃43;" &＃39;hostname&＃39; &＃61; &＃39;hadoop102&＃39;," &＃43;" &＃39;port&＃39; &＃61; &＃39;3306&＃39;," &＃43;" &＃39;username&＃39; &＃61; &＃39;root&＃39;," &＃43;" &＃39;password&＃39; &＃61; &＃39;000000&＃39;," &＃43;" &＃39;database-name&＃39; &＃61; &＃39;test&＃39;," &＃43;" &＃39;table-name&＃39; &＃61; &＃39;z_user_info&＃39;" &＃43;")");tableEnv.executeSql("select * from user_info").print();

自定义反序列化器

public class Flink_CDCWithCustomerSchema {public static void main(String[] args) throws Exception {//1.创建执行环境StreamExecutionEnvironment env &＃61; StreamExecutionEnvironment.getExecutionEnvironment();env.setParallelism(1);//2.创建Flink-MySQL-CDC的SourceProperties properties &＃61; new Properties();//initial (default): Performs an initial snapshot on the monitored database tables upon first startup, and continue to read the latest binlog.//latest-offset: Never to perform snapshot on the monitored database tables upon first startup, just read from the end of the binlog which means only have the changes since the connector was started.//timestamp: Never to perform snapshot on the monitored database tables upon first startup, and directly read binlog from the specified timestamp. The consumer will traverse the binlog from the beginning and ignore change events whose timestamp is smaller than the specified timestamp.//specific-offset: Never to perform snapshot on the monitored database tables upon first startup, and directly read binlog from the specified offset.properties.setProperty("debezium.snapshot.mode", "initial");DebeziumSourceFunction mysqlSource &＃61; MySQLSource.builder().hostname("hadoop102").port(3306).username("root").password("000000").databaseList("test").tableList("test.z_user_info") //可选配置项,如果不指定该参数,则会读取上一个配置下的所有表的数据,注意&＃xff1a;指定的时候需要使用"db.table"的方式.debeziumProperties(properties).deserializer(new DebeziumDeserializationSchema() { //自定义数据解析器&＃64;Overridepublic void deserialize(SourceRecord sourceRecord, Collector collector) throws Exception {//获取主题信息,包含着数据库和表名 mysql_binlog_source.test.z_user_infoString topic &＃61; sourceRecord.topic();String[] arr &＃61; topic.split("\\.");String db &＃61; arr[1];String tableName &＃61; arr[2];//获取操作类型 READ DELETE UPDATE CREATEEnvelope.Operation operation &＃61; Envelope.operationFor(sourceRecord);//获取值信息并转换为Struct类型Struct value &＃61; (Struct) sourceRecord.value();//获取变化后的数据Struct after &＃61; value.getStruct("after");//创建JSON对象用于存储数据信息JSONObject data &＃61; new JSONObject();for (Field field : after.schema().fields()) {Object o &＃61; after.get(field);data.put(field.name(), o);}//创建JSON对象用于封装最终返回值数据信息JSONObject result &＃61; new JSONObject();result.put("operation", operation.toString().toLowerCase());result.put("data", data);result.put("database", db);result.put("table", tableName);//发送数据至下游collector.collect(result.toJSONString());}&＃64;Overridepublic TypeInformation getProducedType() {return TypeInformation.of(String.class);}}).build();//3.使用CDC Source从MySQL读取数据DataStreamSource mysqlDS &＃61; env.addSource(mysqlSource);//4.打印数据mysqlDS.print();//5.执行任务env.execute();}
}