target table has x column(s) but the inserted data has y column(s), including z partition column(s

hive表想使用动态分区的方法，insert overwrite 的时候，必须把动态分区字段放在select 的字段最后面。好多中文帖子写的都不对！坑人

On the other hand, if you used dynamic partitioning -- i.e. INSERT ... PARTITION (country) with no literal value -- then the actual value for the partitioning "column" would have to be fed as an extra column in SELECT, after the real columns.

宝藏来源：hadoop - Inserting into Hive table - Non Partitioned table to Partitioned table - Cannot insert into target table because column number/types - Stack Overflow