hadoop hive sql语句报错

tab0表 是load data local inpath 'x' overwrite into table tab0;
将表中的数据简单清洗写入tab1,
然后查询tab1将结果写入tab4报错,直接tab0不报错。
tab0:

create table tab0(
day_id bigint,
sale_nbr string,
buy_nbr string,
cnt int,
round double
)
row format delimited fields terminated by ','
lines terminated by '\n';

tab1:

create table tab1(
day_id string,
sale_nbr string,
buy_nbr string,
cnt int,
round double
 )
row format delimited fields terminated by ','
lines terminated by '\n';

img

img

img

day_id类型不同吧,一个bigint,一个String

设置成单机模式能运行吗?