flink使用hive作为维表，kafka作为数据源，join时候报错怎么办？

select .... FROM jdqTableSources AS a JOIN tmmmp FOR SYSTEM_TIME AS OF a.proctime AS b

Caused by: org.apache.flink.client.program.ProgramInvocationException: The main method caused an error: Temporal Table Join requires primary key in versioned table, but no primary key can be found. The physical plan is: FlinkLogicalJoin(condition=[AND(=($0, $4), __INITIAL_TEMPORAL_JOIN_CONDITION($3, __TEMPORAL_JOIN_LEFT_KEY($0), __TEMPORAL_JOIN_RIGHT_KEY($4)))], joinType=[inner]) FlinkLogicalCalc(select=[opt, src, cur, PROCTIME() AS proctime]) FlinkLogicalTableSourceScan(table=[[default_catalog, default_database, jdqTableSources]], fields=[mid, db, sch, tab, opt, ts, ddl, err, src, cur, cus]) FlinkLogicalSnapshot(period=[$cor0.proctime]) FlinkLogicalCalc(select=[item_sku_id, premium, cate_lev, type, borc]) FlinkLogicalTableSourceScan(table=[[myhive, dev, dev_brokenscreen_insurance_sku_info]], fields=[item_sku_id, item_sku_name, premium, cate_lev, type, borc, plan_code, subjection_b, product_name, lev_low_price, lev_upp_price, jd_price, shelves_tm, item_first_cate_name, item_second_cate_name, item_third_cate_name, sure_cate_lev, flag])*来自志愿者整理的flink邮件归档