On tis, 2011-01-04 at 22:24 -0500, Robert Haas wrote:
> > Just to be clear: are we saying that "CREATE ROLE foo SUPERUSER"
> > should grant both superuser and replication, as well as the default
> > "postgres" user also having replication as well?
>
> I think that's what we're saying.
So now superusers have it by default but you can explicitly revoke it?
I guess that's still inconsistent with other superuser behavior. You
can't revoke a superuser's CREATEDB bit, for example. (You can, but it
won't have an effect, of course.)