This is most likely caused by data dictionary modification. No data import script should be able to bypass the Unique constraint on the database level.
In my previous post I assumed that there was only one duplicated record. If this is the case a very simple SQL delete statement should do the trick.
delete from tableName
where RecId = 0 – use the RecId of the duplicated record
and DataAreaId = ‘’ – specify the data area ID if applicable
If the delete fails the unique constraint and/or the eprimary index may have to be removed before deleting the duplicated record. In this case it would be a very good idea to backup the table before applying any schema changes 