From owner-freebsd-questions Sat Apr 19 17:56:19 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA25949 for questions-outgoing; Sat, 19 Apr 1997 17:56:19 -0700 (PDT) Received: from who.cdrom.com (who.cdrom.com [204.216.27.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA25929; Sat, 19 Apr 1997 17:56:13 -0700 (PDT) Received: from panda.hilink.com.au (panda.hilink.com.au [203.2.144.5]) by who.cdrom.com (8.8.5/8.6.11) with ESMTP id RAA13222 ; Sat, 19 Apr 1997 17:29:17 -0700 (PDT) Received: (from danny@localhost) by panda.hilink.com.au (8.8.5/8.7.3) id KAA27202; Sun, 20 Apr 1997 10:29:00 +1000 (EST) Date: Sun, 20 Apr 1997 10:28:59 +1000 (EST) From: "Daniel O'Callaghan" To: Chris Samaritoni cc: questions@freebsd.org, isp@freebsd.org Subject: Re: Controlled bandwidth sharing In-Reply-To: <3.0.1.32.19970419125734.009d0c20@gateway> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-questions@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Sat, 19 Apr 1997, Chris Samaritoni wrote: > We have at our disposal for outgoing bandwidth: > > . a router with a dedicated T-1 @ 1.5mbps (flat rate $$) > . a 10baseT switched port on a shared OC-3 (metered rate $$) > > Currently we are on the switched port -- current bandwidth requirements are > about 1mbps sustained and about 8mbps burst. Our provider wants us to move > to a T-1, but we'll lose the nice burst capacity if we move to the T-1. > Ideally, we'd like: > > . first 1.5mbps routed to the T-1 > . additional bandwidth routed to the switch when needed Is the metering inbound only? I would guess that you have more inbound traffic than outbound. If that is the case, you'll have to control the packets' disposition on your feed's side of the link. This is going to be tricky. I have a bandwidth limiter using divert(4) sockets which you might be able to use as a framework to write your own daemon for this. Alternatively, if you keep specific routing info on the FreeBSD box upstream of the link, you can take the code which reads route info packets and bytes and use the info to adjust the destination for particular routes as required, to do the balancing. Another possibility would be to use mpd in ppp-over-tcp mode and hack it to do bandwidth on demand. Danny