Date: Thu, 10 Feb 2000 21:11:22 -0700 From: Wes Peters <wes@softweyr.com> To: Garrett Wollman <wollman@khavrinen.lcs.mit.edu> Cc: Stephen Macmanus <stephenm@windriver.com>, freebsd-net@freebsd.org Subject: Re: Path MTU discovery cutoff Message-ID: <38A38BEA.146E6526@softweyr.com> References: <200002102308.PAA18465@brisbane.wrs.com> <200002110146.UAA44355@khavrinen.lcs.mit.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
Garrett Wollman wrote:
>
> <<On Thu, 10 Feb 2000 15:08:02 -0800 (PST), Stephen Macmanus <stephenm@windriver.com> said:
>
> > RFC 1191 specifies a minimum path MTU estimate of 68 bytes for a
> > participating host. The FreeBSD 3.4 implementation uses a cutoff
> > of 296 bytes. Why does it have a larger minimum?
>
> Because when I felt that the RFC-specified minimum was unreasonably
> small, and if people want to have links with tiny MTUs then they
> should bear the costs themselves. It's also protection against a
> PMTU-DoS attack.
Is this a candidate for a sysctl knob? sys.net.inet.ip.minmtu, with a default
value of 296, so we can be fully RFC compliant and yet sensible?
--
"Where am I, and what am I doing in this handbasket?"
Wes Peters Softweyr LLC
wes@softweyr.com http://softweyr.com/
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-net" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?38A38BEA.146E6526>