Re: BUG #18242: pg_dump with non-superuser from pg14 to pg15 fails on ALTER FUNCTION

From: Aksel Allas <allasaksel(at)gmail(dot)com>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: "Wetmore, Matthew (CTR)" <Matthew(dot)Wetmore(at)express-scripts(dot)com>, Laurenz Albe <laurenz(dot)albe(at)cybertec(dot)at>, pgsql-bugs(at)lists(dot)postgresql(dot)org
Subject: Re: BUG #18242: pg_dump with non-superuser from pg14 to pg15 fails on ALTER FUNCTION
Date: 2023-12-13 06:23:15
Message-ID: CABnLe_C6UbB=mSyDDGHUx3k4=hps1wmehmVmph7=1LG0catT+w@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

In my specific scenario it could be fixed such that e.g. pg_dump has a flag
for non superuser usage.

If the flag is enabled, the dump could check if CREATE privilege is given
at the end of the dump file to the correct schema and instead give the
privilege after connecting to the correct database. Then my case would work.

Best wishes!
Aksel

On Tue, Dec 12, 2023, 6:55 PM Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:

> "Wetmore, Matthew (CTR)" <Matthew(dot)Wetmore(at)express-scripts(dot)com> writes:
> > What about a whole new user type:
> > An 'Admin' account that isn't a super user, but just has dump/ elevated
> permissions /customizable.
>
> [ shrug... ] Maybe, but there's a mighty lot of devils hiding in
> the details. Exactly what special privileges would this user type
> need? How would we convince ourselves (and more to the point,
> convince the cloud providers) that such a set of privileges is
> safe to give out? Poking holes in the privilege model is usually
> a good way to create security hazards.
>
> BTW, please keep the list cc'd.
>
> regards, tom lane
>

In response to

Browse pgsql-bugs by date

  From Date Subject
Next Message Daniel Gustafsson 2023-12-13 08:44:26 Re: BUG #18245: pg_restore accepts same output file name as input file name
Previous Message Tom Lane 2023-12-13 04:04:37 Re: BUG #18244: Corruption in indexes involving whole-row expressions