Flink-1.12支持kafka join jdbc维表吗

请问kafka join jdbc维表数据而不是join jdbc的changelog，支持吗？ 

在给jdbc声明主键和watermark后，使用官网的事件时间join语法，任务启动后发现jdbc算子一次性读取了所有jdbc维表数据，状态变成了finished，这样的话按理来说不管维表数据怎么变kafka都join不到维表数据了呀？ 

CREATE TABLE orders ( 

order_id STRING, 

currency STRING, 

amount INT, 

order_time TIMESTAMP(3), 

WATERMARK FOR order_time AS order_time 

) WITH ( 

'connector' = 'kafka', 

'topic' = 'topic_flink', 

'properties.bootstrap.servers' = '10.3.12.113:9092', 

'properties.group.id' = 'flink', 

'format' = 'json', 

'scan.startup.mode' = 'latest-offset' 

) 

CREATE TABLE latest_rates ( 

currency STRING, 

rate DECIMAL(38, 10), 

currency_time TIMESTAMP(3), 

WATERMARK FOR currency_time AS currency_time, 

PRIMARY KEY (currency) NOT ENFORCED 

) WITH ( 

'connector' = 'jdbc', 

'url' = 'jdbc:mysql://10.3.12.113:3306/base?useUnicode=true&characterEncoding=utf8&serverTimezone=PRC&useSSL=false' 

'username' = 'root', 

'password' = 'root1234', 

'table-name' = 'latest_rates', 

'lookup.cache.max-rows' = '1', 

'lookup.cache.ttl' = '1min' 

) 

SELECT 

o.order_id, 

o.order_time, 

o.amount * r.rate AS amount, 

r.currency 

FROM orders AS o 

LEFT JOIN latest_rates FOR SYSTEM_TIME AS OF o.order_time r 

ON o.currency = r.currency"*来自志愿者整理的flink邮件归档