Re: Backing out of privilege grants rabbit hole

From: raf <raf(at)raf(dot)org>
To: pgsql-general(at)lists(dot)postgresql(dot)org
Subject: Re: Backing out of privilege grants rabbit hole
Date: 2020-04-03 04:57:39
Message-ID: 20200403045739.ej4qsc6cx4nk4yb4@raf.org
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

It's probably more sensible to grant permissions to roles that
represent groups, and have roles for individual users that
inherit the permissions of the group roles. Then you don't
need to revoke the permissions just because an individiual
has left.

cheers,
raf

AC Gomez wrote:

> Thanks for the quick response. The problem is, in most cases the owner is
> not the grantee. So if a role, let's say a temp employee, gets grants, then
> leaves, I can't do a drop owned because that temp never owned those
> objects, he just was granted access. Is there a "drop granted" kind of
> thing?
>
> On Thu, Apr 2, 2020, 11:37 PM Guyren Howe <guyren(at)gmail(dot)com> wrote:
>
> > https://www.postgresql.org/docs/12/sql-drop-owned.html
> >
> > On Apr 2, 2020, at 20:34 , AC Gomez <antklc(at)gmail(dot)com> wrote:
> >
> > Do I understand correctly that if a role was assigned countless object
> > privileges and you want to delete that role you have to sift through a
> > myriad of privilege grants in what amounts to a time consuming trial and
> > error exercise until you've got them all?
> >
> > Or is there a single command that with just delete the role and do a
> > blanket grant removal at the same time?

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message AC Gomez 2020-04-03 04:59:23 Re: Backing out of privilege grants rabbit hole
Previous Message James Brauman 2020-04-03 04:42:39 Improve COPY performance into table with indexes.