From owner-freebsd-hackers Wed Nov 6 14:45:21 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id OAA04036 for hackers-outgoing; Wed, 6 Nov 1996 14:45:21 -0800 (PST) Received: from sumter.awod.com (awod.com [198.81.225.1]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id OAA04031 for ; Wed, 6 Nov 1996 14:45:18 -0800 (PST) Received: from tsunami..awod.com (chs0094.awod.com [198.81.225.93]) by sumter.awod.com (8.7.6/8.6.12) with SMTP id RAA12450; Wed, 6 Nov 1996 17:45:10 -0500 (EST) Message-Id: <3.0.32.19961106173443.0074b1a8@awod.com> X-Sender: klam@awod.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Wed, 06 Nov 1996 17:45:11 -0500 To: Dan Walters , hackers@freebsd.org From: Ken Lam Subject: Re: Limiting bandwidth on a socket? (SO_RCVBUF?) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 10:51 PM 11/5/96 -0600, Dan Walters wrote: >I'm trying to come up with some way to limit the amount of bandwidth on a >socket, so I can have my mail and large files retrieve without slowing >down a telnet session that much. > >I thought I could do this by setting SO_RCVBUF to a small value, but it >doesn't seem to change the window size at all when I look at it with >tcpdump. You need to limit both SNDBUF and RCVBUF since your background xfers are also pushing outgoing :) Perchance that you might drop the buffers to less than the MTU, then you only wait for one background packet ahead of your outbound . Making the buffer even smaller would lessen the wait even further, but would degrade background performance, but decrease latency. Alas, you could always upgrade to ISDN :) -ken --- Ken Lam lam@awod.com Integrated Technical Systems Systems, Networks, and Internet Solutions -- Defining Technology Today "'Plug and Play' was only applicable to the original ATARI(tm)"