| From: | Mike Mascari <mascarm(at)mascari(dot)com> |
|---|---|
| To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
| Cc: | Sander Steffann <sander(at)steffann(dot)nl>, pgsql-general(at)postgresql(dot)org |
| Subject: | Re: Is this a bug, possible security hole, or wrong |
| Date: | 2002-06-13 11:45:37 |
| Message-ID: | 3D0885E1.8F369ACA@mascari.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Tom Lane wrote:
>
> "Sander Steffann" <sander(at)steffann(dot)nl> writes:
> > But he is right in that his trick works. This proves that views can not be
> > safely used for security, which is an important thing to realise...
>
> A different way to look at it is that the privilege of creating
> functions shouldn't be handed out willy-nilly. The trick of hiding
> recording operations in a function can be used in other ways besides
> this one.
Tom,
If a user has permissions to write PL/SQL functions, and the statistics
collector is running with STATS_COMMAND_STRING = true, could not that
user "log" other users' queries using the same technique I described by
querying pg_stat_activity? If so, isn't the labeling of PL/SQL (or
PL/Tcl, PL/Perl, etc.) as 'TRUSTED' an overstatement? ;-)
Mike Mascari
mascarm(at)mascari(dot)com
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2002-06-13 13:06:18 | Re: Help required with installing postgreSQL on SCO Openserver |
| Previous Message | Stephane Bortzmeyer | 2002-06-13 09:49:37 | Setting up the pgReplicator: nightmare in the dark |