Re: @(#)Mordre Labs advisory 0x0005: Several buffer overruns in PostgreSQL

//@(#) Mordred Labs advisory 0x0005

Release data: 23/08/02
Name: Several buffer overruns in PostgreSQL
Versions affected: all versions
Risk: from average to low

--[ Description:

PostgreSQL provides you with several builint geo types
(circle,polygon,box...etc).
Unfortunately the code for geo functions written in a very insecure style
and should be totally rewritten, as a quick search revealed this:

---[ Details:

1)

Upon invoking a polygon(integer, circle) function
a src/backend/utils/adt/geo_ops.c:circle_poly() function will gets called,
which suffers from a buffer overflow.

2) A src/backend/adt/utils/geo_ops.c:path_encode() fails to detect a buffer
overrun condition. It is called in multiple places, the most
interesting are path_out() and poly_out() functions.

3) Upon converting a char string to a path object, a
src/backend/utils/adt/geo_ops.c:path_in() function will gets called,
which suffers from a buffer overrun, caused by a very long argument.

4) A src/backend/utils/adt/geo_ops.c:poly_in() function fails to detect a
buffer
overrun condition caused by a very long argument.

5) A src/backend/utils/adt/geo_ops.c:path_add() also fails to detect a
simple buffer
overrun.

6)

And finally, a truly dumb feature (not a security related though) in
postmaster:
$ postmaster -o `perl -e 'print "\x66" x 1200'`
Segmentation fault (core dumped)

--[ How to reproduce:

I only show how to reproduce a first buffer overrun condition, as the others
too memory consuming :-)

1)

template1=# select polygon(268435455,'((1,2),3)'::circle);
pqReadData() -- backend closed the channel unexpectedly.
This probably means the backend terminated abnormally
before or while processing the request.
The connection to the server was lost. Attempting reset: Failed.
!#

--[ Solution

Drop the vulnerable functions.

________________________________________________________________________
This letter has been delivered unencrypted. We'd like to remind you that
the full protection of e-mail correspondence is provided by S-mail
encryption mechanisms if only both, Sender and Recipient use S-mail.
Register at S-mail.com: http://www.s-mail.com/inf/en

Sir Mordred The Traitor <mordred(at)s-mail(dot)com> writes:
> Upon invoking a polygon(integer, circle) function a
> src/backend/utils/adt/geo_ops.c:circle_poly() function will gets
> called, which suffers from a buffer overflow.
>
> 2) A src/backend/adt/utils/geo_ops.c:path_encode() fails to detect a
> buffer overrun condition. It is called in multiple places, the most
> interesting are path_out() and poly_out() functions.

> 5) A src/backend/utils/adt/geo_ops.c:path_add() also fails to detect
> a simple buffer overrun.

I've attached a patch which should fix these problems.

> 3) Upon converting a char string to a path object, a
> src/backend/utils/adt/geo_ops.c:path_in() function will gets called,
> which suffers from a buffer overrun, caused by a very long argument.

> 4) A src/backend/utils/adt/geo_ops.c:poly_in() function fails to
> detect a buffer overrun condition caused by a very long argument.

I wasn't able to reproduce either of these (wouldn't it require an
input string with several hundred thousand commas?), can you give me a
test-case?

Cheers,

Neil

--
Neil Conway <neilc(at)samurai(dot)com> || PGP Key ID: DB3C29FC

Your patch has been added to the PostgreSQL unapplied patches list at:

http://candle.pha.pa.us/cgi-bin/pgpatches

I will try to apply it within the next 48 hours.

---------------------------------------------------------------------------

Neil Conway wrote:
> Sir Mordred The Traitor <mordred(at)s-mail(dot)com> writes:
> > Upon invoking a polygon(integer, circle) function a
> > src/backend/utils/adt/geo_ops.c:circle_poly() function will gets
> > called, which suffers from a buffer overflow.
> >
> > 2) A src/backend/adt/utils/geo_ops.c:path_encode() fails to detect a
> > buffer overrun condition. It is called in multiple places, the most
> > interesting are path_out() and poly_out() functions.
>
> > 5) A src/backend/utils/adt/geo_ops.c:path_add() also fails to detect
> > a simple buffer overrun.
>
> I've attached a patch which should fix these problems.
>
> > 3) Upon converting a char string to a path object, a
> > src/backend/utils/adt/geo_ops.c:path_in() function will gets called,
> > which suffers from a buffer overrun, caused by a very long argument.
>
> > 4) A src/backend/utils/adt/geo_ops.c:poly_in() function fails to
> > detect a buffer overrun condition caused by a very long argument.
>
> I wasn't able to reproduce either of these (wouldn't it require an
> input string with several hundred thousand commas?), can you give me a
> test-case?
>
> Cheers,
>
> Neil
>
> --
> Neil Conway <neilc(at)samurai(dot)com> || PGP Key ID: DB3C29FC

[ Attachment, skipping... ]

>
> ---------------------------(end of broadcast)---------------------------
> TIP 6: Have you searched our list archives?
>
> http://archives.postgresql.org

--
Bruce Momjian | http://candle.pha.pa.us
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ Christ can be your backup. | Newtown Square, Pennsylvania 19073

Patch applied. Thanks.

---------------------------------------------------------------------------

Neil Conway wrote:
> Sir Mordred The Traitor <mordred(at)s-mail(dot)com> writes:
> > Upon invoking a polygon(integer, circle) function a
> > src/backend/utils/adt/geo_ops.c:circle_poly() function will gets
> > called, which suffers from a buffer overflow.
> >
> > 2) A src/backend/adt/utils/geo_ops.c:path_encode() fails to detect a
> > buffer overrun condition. It is called in multiple places, the most
> > interesting are path_out() and poly_out() functions.
>
> > 5) A src/backend/utils/adt/geo_ops.c:path_add() also fails to detect
> > a simple buffer overrun.
>
> I've attached a patch which should fix these problems.
>
> > 3) Upon converting a char string to a path object, a
> > src/backend/utils/adt/geo_ops.c:path_in() function will gets called,
> > which suffers from a buffer overrun, caused by a very long argument.
>
> > 4) A src/backend/utils/adt/geo_ops.c:poly_in() function fails to
> > detect a buffer overrun condition caused by a very long argument.
>
> I wasn't able to reproduce either of these (wouldn't it require an
> input string with several hundred thousand commas?), can you give me a
> test-case?
>
> Cheers,
>
> Neil
>
> --
> Neil Conway <neilc(at)samurai(dot)com> || PGP Key ID: DB3C29FC

[ Attachment, skipping... ]

>
> ---------------------------(end of broadcast)---------------------------
> TIP 6: Have you searched our list archives?
>
> http://archives.postgresql.org

--
Bruce Momjian | http://candle.pha.pa.us
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ Christ can be your backup. | Newtown Square, Pennsylvania 19073