From owner-freebsd-questions  Mon Dec 25 16:21:19 1995
Return-Path: owner-questions
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id QAA08750
          for questions-outgoing; Mon, 25 Dec 1995 16:21:19 -0800 (PST)
Received: from Root.COM (implode.Root.COM [198.145.90.17])
          by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id QAA08745
          Mon, 25 Dec 1995 16:21:16 -0800 (PST)
Received: from corbin.Root.COM (corbin [198.145.90.50]) by Root.COM (8.6.12/8.6.5) with ESMTP id QAA07053; Mon, 25 Dec 1995 16:21:03 -0800
Received: from localhost (localhost [127.0.0.1]) by corbin.Root.COM (8.7.3/8.6.5) with SMTP id QAA05890; Mon, 25 Dec 1995 16:21:04 -0800 (PST)
Message-Id: <199512260021.QAA05890@corbin.Root.COM>
To: Petri Helenius <pete@sms.fi>
cc: kelly@fsl.noaa.gov (Sean Kelly), questions@freebsd.org,
        wollman@freebsd.org
Subject: Re: Unpleasant scrolling behaviour on 2.1.0 (fwd) 
In-reply-to: Your message of "Mon, 25 Dec 95 23:27:28 +0200."
             <199512252127.XAA26484@silver.sms.fi> 
From: David Greenman <davidg@Root.COM>
Reply-To: davidg@Root.COM
Date: Mon, 25 Dec 1995 16:21:02 -0800
Sender: owner-questions@freebsd.org
Precedence: bulk

>Sean Kelly writes:
>
> > Okay, the magic 1500 works ... I suppose if I thought about it, that'd
> > be the correct size for a full IP header plus checksum or other such
> > explanation, but being Christmas day and all, I'll just assume a
> > little magic. :-)
> >
>No it's just that your MTU should be smaller than your TCP window size,
>othervise you make TCP a stop-and-go protocol. I have no idea when the
>non-networking skilled individual that changed the lo0 MTU default to
>16384 is going to correct his mistake. It's an obvious annoyance to
>everyone who's using 127.0.0.1 for any purpose.

   The loopback MTU has been at 16K or greater since Garrett Wollman first
changed it to 64k back in FreeBSD 1.1. I changed it from 64K to 16K back in
March of this year because of some bad interaction that 64K has with the
socket layer. As for "stop-and-go", no, having the MTU at 16k doesn't by
itself cause this - especially since the window size is significantly larger
than that. You can use a variety of network tools to verify this.
   It should probably be lowered because of bugs in the networking code. This
has likely started to become a problem recently because of some bugs fixes
in areas that indirectly affect the Nagel algorithm. 'ttcp -t localhost' still
gives very good results for the 16K MTU, so it's not completely broken.
   It's very presumptuous to assume that "non-networking skilled individuals"
were involved in this change. Quite the contrary. If you want to be helpful,
then I'd suggest making useful comments about what you think the MTU should be
changed to and why you think it should be changed. Your comments will
otherwise only alienante you and cause people to ignore you.

-DG