Re: public schema default ACL

From: Noah Misch <noah(at)leadboat(dot)com>
To: Joe Conway <mail(at)joeconway(dot)com>
Cc: pgsql-hackers(at)postgresql(dot)org
Subject: Re: public schema default ACL
Date: 2018-03-06 07:44:51
Message-ID: 20180306074451.GB1926201@rfd.leadboat.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Sat, Mar 03, 2018 at 02:31:58AM -0800, Joe Conway wrote:
> On 03/03/2018 01:56 AM, Noah Misch wrote:
> > If we do that alone, databases reaching v11 via dump/reload or pg_upgrade will
> > get the new default ACL if they had not changed the ACL of schema public. If
> > they had GRANTed or REVOKEd on schema public, pg_dump will recreate the
> > resulting ACL. This is the standard pg_dump behavior for ACLs on system
> > objects. I think that's okay for the public schema, too, and I like
> > preserving that usual rule. However, if we wanted to minimize upgrade-time
> > surprises, we could make pg_dump include GRANT for schema public
> > unconditionally. That way, the default ACL change would apply to new
> > databases only. Does anyone want to argue for that?
>
> What about a pg_dump option to do that and then a big note in the
> release notes telling people why they might want to use it?

I'd want any new pg_dump option to have use beyond this one case. That is,
not --old-public-schema-acl, but perhaps --old-system-acls-for=OBJECT-PATTERN.
But it's a simple task to loop over your databases and run a GRANT, so I
somewhat doubt that particular idea should win. Hmm.

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Ildar Musin 2018-03-06 07:57:27 Re: using index or check in ALTER TABLE SET NOT NULL
Previous Message Edmund Horner 2018-03-06 07:31:31 Re: PATCH: psql tab completion for SELECT