Hive and Impala duplicating rows when INSERT OVERWRITE

[#1]

Hi there!!!

I am using Hadoop on Cloudera and I was trying to do an INSERT OVERWRITE of a table to another.

The problem:

Table1 has 1561 rows

INSERT INTO TABLE Table2 SELECT * from Table1;

After this I have 1715 rows on Table2.

I have also tried to create table2 as:

CREATE TABLE Table2 AS SELECT * from Table1;

After this I have 1715 rows on Table 2 as well.

I have checked and the different values are all duplicates

select count(*)
from Table2
group by all_columns
having count(*) > 1;

154 duplicated rows.

Why is this happening?

I have tried these using Hive and Impala.

Can someone help me to solve this?

[#2]

I can't really conceive of any explanation other than that the duplicate rows existed in the original table.