You are viewing a plain text version of this content. The canonical link for it is here.

Posted to user-zh@flink.apache.org by "kandy.wang" <ka...@163.com> on 2020/11/30 14:50:04 UTC

flink cdc 如何保证group agg结果正确性

insert into kudu.default_database.index_agg
SELECT v_spu_id as spu_id,sum(leaving_num*vipshop_price) as leaving_price,DATE_FORMAT(LOCALTIMESTAMP,'yyyy-MM-dd HH:mm:ss')
FROM  XX.XX.XX
group by v_spu_id;


XX.XX.XX 是通过自定义cdc format消费公司的cdc数据源，cdc数据源在kafka，数据只保留7天数据，都是增量消费，如何保证结果准确。
怎么做初始化，这个初始化，是把数据初始化到state里么？ 现在通过对数发现，数据量对不上。