flinkcdc使用笔记

本文主要包括：

• flinkcdc使用笔记

flinkcdc使用笔记

flink-cdc 3.4可以支持pipeline写数据到iceberg，这里尝试一下

安装

需要注意的是，flink-cdc3.4需要使用最低1.19版本的flink

1. 解压flink和flink-cdc的安装包

   tar zxvf flink-1.19.2-bin-scala_2.12.tgz -C . 
   tar zxvf flink-cdc-3.4.0-bin.tar.gz -C .

2. 下载pipeline包

   cp flink-cdc-pipeline-connector-iceberg-3.4.0.jar  flink-cdc-3.4.0/lib
   cp flink-cdc-pipeline-connector-mysql-3.4.0.jar  flink-cdc-3.4.0/lib
   cp mysql-connector-j-8.0.31.jar flink-cdc-3.4.0/lib

3. 下载iceberg runtime包

   cp flink-sql-connector-hive-2.3.9_2.12-1.17.2.jar flink-1.19.2/lib
   cp iceberg-flink-runtime-1.17-1.4.3.jar flink-1.19.2/lib
   cp mysql-connector-j-8.0.31.jar flink-1.19.2/lib

4. 启动flink standalone集群

   bin/start-cluster.sh

5. 编写mysql-to-iceberg.yaml

   source:
     type: mysql
     name: MySQL Source
     hostname: 172.16.2.205
     port: 3306
     username: root
     password: root123456
     tables: test.gjc_test_iceberg_streaming_delay
     server-id: 5401-5404
   
   sink:
     type: iceberg
     name: Iceberg Sink
     catalog.properties.type: hive
     catalog.properties.uri: thrift://172.16.2.204:9083
     catalog.properties.hive-conf-dir: /root/gujc/flink-1.17.2/conf
   
   pipeline:
     name: MySQL to Iceberg Pipeline
     parallelism: 1

6. 启动flink-cdc任务

   bin/flink-cdc.sh mysql-to-iceberg.yaml 

   数据可以正常采集进iceberg，但是，和手动写flinksql实时采集写icebebrg效果一样，没办法捕捉增量数据
   flink-cdc自动建表的语句比较简单：

   | CREATE TABLE `iceberg_catalog`.`test`.`gjc_test_iceberg_streaming_delay` (
     `id` INT NOT NULL,
     `name` VARCHAR(2147483647),
     `age` BIGINT,
     `create_time` TIMESTAMP(6) WITH LOCAL TIME ZONE NOT NULL,
     CONSTRAINT `c7fdfd2e-19ca-418f-9ab2-a9458a7f32f9` PRIMARY KEY (`id`) NOT ENFORCED
   ) WITH (
     'write.parquet.compression-codec' = 'zstd'
   )

   通过以下sql建表，可以实时读取iceberg表，但是没办法区分是insert还是update还是delete

   CREATE TABLE iceberg_catalog.iceberg.gjc_test_iceberg_streaming_delay (
     id int NOT NULL,
     name STRING ,
     age int ,
     create_time TIMESTAMP ,
     update_time TIMESTAMP,
     PRIMARY KEY(`id`) NOT ENFORCED
   ) WITH (
     'format-version' = '2',
     'write.format.default' = 'parquet',
     'write.parquet.compression-codec' = 'uncompressed',
     'write.delete.parquet.compression-codec' = 'uncompressed',
     'write.delete.mode'='merge-on-read',
     'write.merge.mode'='merge-on-read',
     'write.update.mode'='merge-on-read',
     'table.retention.max-snapshots' = '5',
     'write.metadata.delete-after-commit.enabled'='true',
     'write.metadata.previous-versions-max'='5'
   );

   通过以下sql建表，可以实时读取iceberg表，但是读取insert数据

   CREATE TABLE iceberg_catalog.iceberg.gjc_test_iceberg_streaming_delay (
     id int NOT NULL,
     name STRING ,
     age int ,
     create_time TIMESTAMP ,
     update_time TIMESTAMP,
     PRIMARY KEY(`id`) NOT ENFORCED
   ) WITH (
     'format-version' = '2',
     'write.format.default' = 'parquet',
     'write.parquet.compression-codec' = 'uncompressed',
     'write.delete.parquet.compression-codec' = 'uncompressed',
     'write.delete.mode'='merge-on-read',
     'write.merge.mode'='merge-on-read',
     'write.update.mode'='merge-on-read',
     'table.retention.max-snapshots' = '5',
     'write.metadata.delete-after-commit.enabled'='true',
     'write.metadata.previous-versions-max'='5'
   );

   通过以下sql建表，可以无法实时读取iceberg表

   CREATE TABLE iceberg_catalog.iceberg.gjc_test_iceberg_streaming_delay (
     id int NOT NULL,
     name STRING ,
     age int ,
     create_time TIMESTAMP ,
     update_time TIMESTAMP,
     PRIMARY KEY(`id`) NOT ENFORCED
   ) WITH (
     'format-version' = '2',
     'write.upsert.enabled' = 'true',
     'write.format.default' = 'parquet',
     'write.parquet.compression-codec' = 'uncompressed',
     'write.delete.parquet.compression-codec' = 'uncompressed',
     'write.delete.mode'='merge-on-read',
     'write.merge.mode'='merge-on-read',
     'write.update.mode'='merge-on-read',
     'table.retention.max-snapshots' = '5',
     'write.metadata.delete-after-commit.enabled'='true',
     'write.metadata.previous-versions-max'='5'
   );