[76809] in North American Network Operators' Group
Re: Smallest Transit MTU
daemon@ATHENA.MIT.EDU (Robert E.Seastrom)
Thu Dec 30 01:01:53 2004
To: Dan Hollis <goemon@anime.net>
Cc: Florian Weimer <fw@deneb.enyo.de>,
Jerry Pasker <info@n-connect.net>, <nanog@merit.edu>
From: Robert E.Seastrom <rs@seastrom.com>
Date: Thu, 30 Dec 2004 01:00:22 -0500
In-Reply-To: <Pine.LNX.4.44.0412291532180.28453-100000@sasami.anime.net> (Dan
Hollis's message of "Wed, 29 Dec 2004 15:39:28 -0800 (PST)")
Errors-To: owner-nanog-outgoing@merit.edu
Dan Hollis <goemon@anime.net> writes:
>> Why is this a problem? ECN has to be deployed on routers, and it
>> currently isn't.
>
> Because tcp connection endpoints have to implement ECN in order to manage
> the flow.
A naive reader might think from Dan's posting that the Internet didn't
work at all before ECN was codified (experimental with RFC 2481 in
January 1999 and standards-track with RFC 3168 in September 2001).
Heck, if I didn't know better, I'd think he was trolling.
Of course, I may just be a grouchy malcontent who doesn't use RED, any
other kind of active queue management, or take hits from the
end-to-end-QoS-bong when it is passed in my direction... since the
only QoS for which I have any use is Quantity of Service. ECN has
always looked to me like a solution in search of a problem, which may
be why so few people have their panties in a bunch over non-support of
it.
---Rob
