I'm going to create a UNIQUE key in a large, old table (several hundred thousand rows) to enforce each row containing unique data. I can't just up and create the index though, because the table already contains some duplicates. I don't know how many duplicates it contains, or if the duplicates are necessary. In short, I can't analyse the problem before I can come up with a query that selects all the rows that would violate the future UNIQUE key.
Say I have a table with the fields col_one, col_two, col_three and col_four. My future UNIQUE key will contain col_two and col_three. How do I select all the rows in the table where col_two and col_three has a duplicate in the same table?