From owner-freebsd-chat  Sun Feb 16 21:25:49 1997
Return-Path: <owner-chat>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id VAA24914
          for chat-outgoing; Sun, 16 Feb 1997 21:25:49 -0800 (PST)
Received: from darkstar (ras514.srv.net [205.180.127.14])
          by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id VAA24908
          for <chat@FreeBSD.ORG>; Sun, 16 Feb 1997 21:25:43 -0800 (PST)
Received: (from cmott@localhost) by darkstar (8.6.12/8.6.12) id WAA01735; Sun, 16 Feb 1997 22:25:30 -0700
Date: Sun, 16 Feb 1997 22:25:28 -0700 (MST)
From: Charles Mott <cmott@srv.net>
X-Sender: cmott@darkstar
To: chat@FreeBSD.ORG
Subject: Re: CRL <--> MCI at pacbell NAP 
In-Reply-To: <199702170444.UAA04626@root.com>
Message-ID: <Pine.BSF.3.91.970216221508.1692B-100000@darkstar>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-chat@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

On Sun, 16 Feb 1997, David Greenman wrote:
> >Part of a traceroute to wcarchive:
> >
> > 9  pacbell-nap-atm.SanFrancisco.mci.net (204.70.1.202)  48.998 ms  50.476 ms  49.367 ms
> >10  pacbell-nap-atm.SanFrancisco.mci.net (204.70.1.202)  53.675 ms  47.162 ms  61.408 ms
> >11  pb-nap.crl.net (198.32.128.20)  1513.402 ms * *
> >12  * wcarchive.cdrom.com (165.113.58.253)  483.625 ms *
> >
> >Not so hot; can't get more than 1 kbyte/sec from wcarchive when normally
> >I can get over 200.  <sigh>
> >
> >Anyone have any idea whose fault it is (looks like CRL to me, but could
> >be something else at the NAP) and when it will be fixed?  Guess I could
> >manually route around it but that's annoying.
> 
>    Yes, it's MCI's fault. Their connection to the PB-NAP is overloaded (by at
> least a factor of 2 as near as I can tell). CRL has been bugging them about it,
> but the word from MCI is that there is no plan to upgrade their connection
> to the NAP to OC3 (155Mbps); I believe it is currently only 34Mbps, but might
> be DS3. It appears that this policy is an attempt to force larger service
> providers into doing 'private' peering (i.e. via a private point-to-point
> connection between the providers, bypassing the NAP).

It appears that MCI is pushing some of the routing burden back towards the
service providers.  This seems to be the direction a lot of networking is
taking.  I wonder if their goal is forcing customers to buy more bandwidth
than they really need by making it less efficiently used. 

Not to be too trite here, but there is an old saying that when elephants
fight, the grass beneath their feet suffers.  1 kbyte/sec is pretty 
slow.

Charles Mott