From owner-freebsd-questions  Mon Aug  5 15:05:28 1996
Return-Path: owner-questions
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id PAA25617
          for questions-outgoing; Mon, 5 Aug 1996 15:05:28 -0700 (PDT)
Received: from mail.think.com (Mail1.Think.COM [131.239.33.245])
          by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id PAA25609
          for <questions@freebsd.org>; Mon, 5 Aug 1996 15:05:26 -0700 (PDT)
Received: from Early-Bird.Think.COM (Early-Bird-1.Think.COM [131.239.146.105])
    by mail.think.com (8.7.5/m3) with ESMTP id SAA22968
    for <questions@freebsd.org>; Mon, 5 Aug 1996 18:05:23 -0400 (EDT)
Received: from compound.Think.COM ([206.147.16.34])
    by Early-Bird.Think.COM (8.7.5/e1) with ESMTP id SAA06583
    for <questions@freebsd.org>; Mon, 5 Aug 1996 18:05:05 -0400 (EDT)
Received: (from alk@localhost) by compound.Think.COM (8.7.5/8.7.3) id RAA25880; Mon, 5 Aug 1996 17:05:06 -0500 (CDT)
Date: Mon, 5 Aug 1996 17:05:06 -0500 (CDT)
From: Tony Kimball <alk@think.com>
Message-Id: <199608052205.RAA25880@compound.Think.COM>
To: questions@freebsd.org
Subject: tcp tuning
Sender: owner-questions@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk


A question for the IP stack experts: I use the Internet quite a lot
between Chicago and Boston, on the MCI backbone.  MCI drops 25-50% of
the packets at Willow Springs, and has done for weeks now.  This leads
to absolutely terrible telnet latencies -- unusable, even.  Is there a
way to tune the TCP timing parameters to provide optimal latency under
lossy conditions, perhaps at the expense of bandwidth.  Retrying more
rapidly, for example, would be a help.  What else would be helpful?