netdev - Re: ipsec: ipcomp alg problem on vti interface

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20171127131153.dc5oml6fypyqf3yh@gauss3.secunet.de>
Date:   Mon, 27 Nov 2017 14:11:53 +0100
From:   Steffen Klassert <steffen.klassert@...unet.com>
To:     Alexey Kodanev <alexey.kodanev@...cle.com>
CC:     Petr Vorel <pvorel@...e.cz>, netdev <netdev@...r.kernel.org>
Subject: Re: ipsec: ipcomp alg problem on vti interface

On Mon, Nov 27, 2017 at 04:00:38PM +0300, Alexey Kodanev wrote:
> On 11/27/2017 03:07 PM, Steffen Klassert wrote:
> > On Wed, Nov 22, 2017 at 07:06:13PM +0300, Alexey Kodanev wrote:
> >>
> >> Is there some flaw in setup or vti not designed to handle ipcomp alg that
> >> can send packets with/without compression (or without further encryption)?
> > VTI is not designed to handle packets without IPsec header, so yes
> > this does not work well with ipcomp that might omit the compression
> > header.
> 
> if so, is it reasonable to keep ipcomp handling in VTI?

I'd say it does not make sense to use vti with ipcomp, but it might
be that somebody created a usecase from that. Unfortunately it is
always critical to remove functionality that is exposed to userspace,
even if it is broken.