Subject: Re: in-kernel pppoe success !
To: None <current-users@netbsd.org>
From: Hal Snyder <hal@vailsys.com>
List: current-users
Date: 11/01/2001 14:23:09
Martin Husemann <martin@duskware.de> writes:

> > Actually, I was thinking it belongs in ip_output.c, with a switch
> > (-proto3?) that could be set per route to cap the TCP MSS at the
> > MTU for that route (or interface if there is no MTU for the route)
> > for sessions initiated external to the NetBSD box.
> 
> I have some (more philosophical) problems with MSS clamping per se,
> but since it seems to be usefull for some people (and provided it's
> optional and off by default) I can cope ;-)
> 
> Anyway, please publish this code (cleaned up or not) and run this
> through the tech-net@netbsd.org mailing list!

Indeed - something like this has been on our to-do list for a month or
so, giving similar capability to what some dedicated routers offer (am
thinking of Cisco "ip tcp adjust-mss nnn", etc.). All we would need is
a per-interface (as opposed to per-route) clamp on traffic going out a
given (gif) interface. Definitely interested!