Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 16 Aug 2018 13:58:07 +0100
From:      "Kristof Provost" <kp@FreeBSD.org>
To:        "Alexandr Krivulya" <shuriku@shurik.kiev.ua>
Cc:        freebsd-security@freebsd.org
Subject:   Re: FreeBSD Security Advisory FreeBSD-SA-18:08.tcp
Message-ID:  <5BB8F247-B799-4839-9E0E-E331B8EA85DB@FreeBSD.org>
In-Reply-To: <306fd368-1093-ace2-7075-a9c6d2bf6860@shurik.kiev.ua>
References:  <20180815054732.9D8C61C2C8@freefall.freebsd.org> <306fd368-1093-ace2-7075-a9c6d2bf6860@shurik.kiev.ua>

next in thread | previous in thread | raw e-mail | index | archive | help
On 15 Aug 2018, at 15:25, Alexandr Krivulya wrote:
> Hi, freebsd-security
>
> Can CVE-2018-6922 be addressed by pf's  fragment reassemble and 
> reassemble tcp options or can it potentially lead to memory overflow 
> (set limit frags?) when this options enabled?
>
No. While pf does limit the maximum number of IP fragments it’ll hold 
on to, this number is large enough that it’s still possible to cause 
the it to use excessive amounts of CPU time.

pf does not reassemble tcp segments, so it won’t protect you agains 
that variant of the attack. The good news there is that it is not itself 
vulnerable to it (for the same reason).

I’m looking into limiting the number of fragments per packet to ensure 
there can’t be excessive CPU use, but that’s not ready to be 
committed yet.

—
Kristof


Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?5BB8F247-B799-4839-9E0E-E331B8EA85DB>