Re: privileges oddity

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Scott Ribe <scott_ribe(at)elevated-dev(dot)com>
Cc: Adrian Klaver <adrian(dot)klaver(at)aklaver(dot)com>, PostgreSQL General <pgsql-general(at)lists(dot)postgresql(dot)org>
Subject: Re: privileges oddity
Date: 2020-08-07 20:38:31
Message-ID: 2798112.1596832711@sss.pgh.pa.us
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-general

Scott Ribe <scott_ribe(at)elevated-dev(dot)com> writes:
> So, one last follow-up, perhaps \du or \du+ should show when a role is mapped that way. If I'd seen a clue to this setting that had been made "before I got here" it would have been figured out sooner.

\drds does already show this; of course, you have to know to look at it,
but the same could be said of \du ...

> I realize ALTER ROLE... SET... can be used to set many more defaults, and there could be some debate about how much to display with \du[+], but the fact that a role abandons all its privs and adopts a different set seems like pretty important info to surface ;-)

IIRC, you aren't the first to get burnt this way. I've wondered for some
time if we shouldn't forbid certain GUCs from being set via ALTER ROLE or
ALTER DATABASE. "role" and "session authorization" are the poster
children here but there might be others. On the other hand, if we do so
somebody will likely complain that they have a legit use-case for it.

regards, tom lane

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Condor 2020-08-09 06:12:14 Need help how to reproduce MySQL binary to PosgreSQL
Previous Message Scott Ribe 2020-08-07 20:29:29 Re: privileges oddity