<Glitchbane@gmail.com> wrote in message
news:1187131725.743277.120860@i38g2000prf.googlegroups.com...
>I have set up a simple transactional replication that includes
> specified columns in one table. One of those columns is used in a
> unique key with the ignore_dup_key option set to 'on'. When I tried
> to get the replication started, it fails with this error:
> 'IGNORE_DUP_KEY' is not a recognized ALTER TABLE option.
>
> I looked at the generated script that creates the unique key that is
> on one of the columns of the table. Although the key on the original
> table was scripted as ... WITH IGNORE_DUP_KEY=ON... the script that
> SQL generates for the replication's initial snapshot is this:
>
> ALTER TABLE [dbo].[PostHeap] ADD CONSTRAINT [PK_PostHeap_new] PRIMARY
> KEY CLUSTERED ([PostID]) WITH IGNORE_DUP_KEY
> go
>
> Thus the error gets generated.
>
> My question is, how can I get around this? I can't just change the
> script because when I restart the synchronization, the script is
> regenerated.
>
> All help greatly appreciated,
>
> Sheila
>

On Aug 15, 5:59 am, "Hilary Cotter" <hilary.cot...@gmail.com> wrote:
> Could you use a post snapshot script to alter the table for this?
>
> Note that I don't really understand why you are using this as if the
> subscriber is a mirror of the publisher this constraint will be enforced at
> the publisher and duplicate values will never come from the publisher to the
> subscriber.
>
> --