Question

Is possible to add a primary key in a table that already have a unique nonclustered index?

I have a table cargo_car and i need to define cd_cargo_car as a PK.

I tried this:

ALTER TABLE dbo.cargo_car ADD PRIMARY KEY (cd_cargo_car)

but i received the error:

Error (1921) An index with the same columns inthe same order alredy exists onthe table

This table has many dependences and when i try to drop the index I receive:

Error (3712) Cannot drop index 'cargo_car.XPKcargo_car' because it still has referential integrity constraints.

This is the Create script:

CREATE TABLE dbo.cargo_car
    (
    cd_cargo_car      SMALLINT NOT NULL,
    nm_cargo_car      VARCHAR (40) NOT NULL,
    cd_refini_car     CHAR (4) NOT NULL,
    cd_reffin_car     CHAR (4) NOT NULL,
    cd_jornada1_car   CHAR (2) NOT NULL,
    cd_jornada2_car   CHAR (2) NOT NULL

    )
GO

CREATE UNIQUE NONCLUSTERED INDEX XPKcargo_car
    ON dbo.cargo_car  (cd_cargo_car)
GO

Suggestion on how to do this?

Tks

Answer 1

you’re going to have to drop all the foreign key contraints on the
other tables, drop that index, create the Primary Key (I guess you
want it to be clustered?), then re-create the foreign key constraints.
– Dems

To allow closing this question, I converted that comment to an answer. It is correct and similar to what I wanted to write…