Subject: Re: silo and packets: overflowing too-bigness
To: None <perry@piermont.com>
From: Bill Sommerfeld <sommerfeld@orchard.arlington.ma.us>
List: port-mac68k
Date: 07/29/1998 16:35:44
> It turns off the TCP long window support that some counterparties
> don't like. I don't know what impacts that might have, but the
> original poster seems to think that will help.

Some implementations of TCP header compression in SLIP and PPP
reportedly break when this option is in use (causing packets to be
corrupted upon decompression).  Given the way VJ header compression
works -- retransmissions aren't compressed -- this merely causes
*severe* performance degradation (albeit down below the level of most
people's patience).



					- Bill