[2912] in linux-net channel archive

home help back first fref pref prev next nref lref last post

Re: [masq] IP masquerading and fragmentation

daemon@ATHENA.MIT.EDU (Graham Mitchell)
Wed May 15 02:44:23 1996

From: "Graham Mitchell" <gmitch@woodlea.wintermute.co.uk>
To: Nigel Metheringham <Nigel.Metheringham@theplanet.net>,
        linux-net@vger.rutgers.edu, masq@lists.indyramp.com, alan@cymru.net
Date: 	Tue, 14 May 1996 21:41:31 +0000
Reply-to: gmitch@woodlea.wintermute.co.uk

On 14 May 96 at 17:26, masq@indyramp.com wrote:

> I have been running some tests on the IP Masquerading stuff, 
> generally with reasonable results at present.  However there is one 
> case that completely floors it...
> 
> 
> The situation is:-
> 	<A> a client box, with reasonable IP stack that does
> 	    (or attempts) MTU discovery - ie a Linux box
> 	    It communicates with the internet via...
> 
> 	<B> a Linux box acting as a masquerading router.
> 	    The private network is one side, with <A> on it,
> 	    the internet (by some means) on the other.

EXACTLY the setup I have.....:) Thank goodness someone else has 
managed to show up the problem. I was beginning to think I was living 
in the Twilight Zone here.....:)) In my case the router with the low 
MTU is my IAP. The force it to 296, and wont change it, 'cos it works 
fine for everyone else, and we dont want to risk screwing them up for 
you', which I see their point if they're in business - doesnt help me 
much tho.....:((

As you say, stripping the DF flag is quick, dirty and 'the wrong 
thing'. However, if someone can point me to where it should be done, 
I'll do my sources (I'm not intimate enough with the IPv4 code to be 
100% sure that I'd do it correctly)

Also, since I dont know the code well enough to fix it, I will leave 
it to those others with the knowledge. I am however MORE than willing 
to test any patches anyone comes up with.

Graham

Ask not what you can do for your country,
but what your government is doing to you


home help back first fref pref prev next nref lref last post