[<prev] [next>] [<thread-prev] [month] [year] [list]
Date: Tue, 12 May 2009 16:18:42 +0200
From: Tomas Hoger <thoger@...hat.com>
To: oss-security@...ts.openwall.com, coley@...re.org
Subject: Re: ipsec-tools 0.7.2
On Wed, 29 Apr 2009 16:56:58 +0200 Tomas Hoger <thoger@...hat.com>
wrote:
> * src/racoon/crypto_openssl.c: From Stephen Bevan: Fix a x509
> signature verification memory leak.
>
> https://trac.ipsec-tools.net/ticket/303
> http://cvsweb.netbsd.org/bsdweb.cgi/src/crypto/dist/ipsec-tools/src/racoon/crypto_openssl.c.diff?r1=1.11.6.4&r2=1.11.6.5&f=h
>
> This leak occurs during user authentication using certificates. It's
> possible to reach it for unauthenticated users, though certificate
> itself is validated first, which mitigates this slightly.
>
> * src/racoon/nattraversal.c: Fix a memory leak in nat-t keepalive
> code.
>
> http://cvsweb.netbsd.org/bsdweb.cgi/src/crypto/dist/ipsec-tools/src/racoon/nattraversal.c.diff?r1=1.6&r2=1.6.6.1&f=h
>
> This can occur during phase1 too, before authentication. Requires
> nat-t to be enabled / allowed, leaks two struct sockaddr.
I'm bit unsure about how to treat these form CVE point of view. These
both happen during normal operation too. However, attacker can cause
these leaks in some setups (ipsec server serving road warriors) without
being able to authenticate successfully, so this bears some exploitation
potential. Given the previous ipsec-tools CVE assignments
(CVE-2008-3651/2), this may deserve CVE too.
Thoughts?
--
Tomas Hoger / Red Hat Security Response Team
Please check out the
Open Source Software Security Wiki, which is counterpart to this
mailing list.
Hosted by DataForce ISP -
Powered by Openwall GNU/*/Linux