From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Bruno Wolff III <bruno(at)wolff(dot)to> |
Cc: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: Problem granting usage on language c |
Date: | 2002-10-20 17:28:31 |
Message-ID: | 13143.1035134911@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general pgsql-hackers |
Bruno Wolff III <bruno(at)wolff(dot)to> writes:
> When I try to grant usage on language c in 7.3b1 I get an error message
> saying I can't because the language isn't trusted. Is this different in b2?
No.
> I can see using whether or not a language is trusted to set defaults,
> but now that access can be controlled it would be useful to grant
> access to load c functions to users that are trusted.
There is no point at all in providing shades of access to untrusted
languages. A user who can define his own C functions is God: he can
use the functions to bypass any security restrictions you may think
you have on him. So it's useless to say that any privilege state
less than superuser-dom is appropriate for creating C functions.
later ...
> The risk is that the admins have to realize that granting access to
> untrusted languages is giving away the keys to the castle.
Equating it to superuser status seems an effective way of reminding
them of that.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2002-10-20 17:37:01 | Re: Using current_database() in a grant |
Previous Message | Bruno Wolff III | 2002-10-20 16:25:34 | Re: Problem granting usage on language c |
From | Date | Subject | |
---|---|---|---|
Next Message | Hunter Hillegas | 2002-10-20 19:05:04 | Buffers and MacOS X |
Previous Message | Bruno Wolff III | 2002-10-20 16:25:34 | Re: Problem granting usage on language c |