Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 21 Jun 2000 23:59:04 -0400
From:      "Louis A. Mamakos" <louie@TransSys.COM>
To:        Garrett Wollman <wollman@khavrinen.lcs.mit.edu>
Cc:        freebsd-net@FreeBSD.ORG
Subject:   Re: Using timestamp option of ip header (IPOPT_TS) 
Message-ID:  <200006220359.XAA35435@whizzo.transsys.com>
In-Reply-To: Your message of "Wed, 21 Jun 2000 23:43:40 EDT." <200006220343.XAA42436@khavrinen.lcs.mit.edu> 
References:  <Pine.GSO.4.21.0006210858450.22741-100000@jungle.owlnet.rice.edu> <200006220225.WAA35025@whizzo.transsys.com> <200006220343.XAA42436@khavrinen.lcs.mit.edu> 

next in thread | previous in thread | raw e-mail | index | archive | help
> <<On Wed, 21 Jun 2000 22:25:45 -0400, "Louis A. Mamakos" <louie@TransSys.COM> said:
> 
> > I don't think you want any IP options present at all.  Depending on the
> > specific implementations in routers, some (most?) will punt IP datagrams
> > with options to a conventional CPU to process.
> 
> It's even worse than that... some routers will punt any packets which:
> 
> 	- are fragmented
> 	- have options
> 	- don't have the ``right'' protocol field
> 
> The second is well-known; I've heard of the first and have actually
> measured the third. 

Yeah, the second became well-known in the early MBONE days where 
source routed packets were used before IP-in-IP tunnels were used to
construct the virtual topology.

But the first and the third are just broken.  I'm not surprised (little
surprises me any more), but mostly disappointed.  I'll have to make
sure our lab guys add these things to their performance tests.

louie








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?200006220359.XAA35435>