From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from tuna.sandelman.ca (tuna.sandelman.ca [IPv6:2607:f0b0:f:3:216:3eff:fe7c:d1f3]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id A42AB3B2B3; Mon, 18 Jan 2016 20:06:28 -0500 (EST) Received: from sandelman.ca (obiwan.sandelman.ca [209.87.249.21]) by tuna.sandelman.ca (Postfix) with ESMTP id E052F2002A; Mon, 18 Jan 2016 20:14:23 -0500 (EST) Received: from obiwan.sandelman.ca (ip6-localhost [IPv6:::1]) by sandelman.ca (Postfix) with ESMTP id 4F8FE637A0; Mon, 18 Jan 2016 20:06:26 -0500 (EST) From: Michael Richardson To: Jonathan Morton cc: Valent Turkovic , "cerowrt-devel\@lists.bufferbloat.net" , make-wifi-fast@lists.bufferbloat.net In-Reply-To: <6ACF3DA5-AEA7-43DD-A3BA-33F680374F09@gmail.com> References: <6737994D-CE0B-46F5-B55C-A584FF6A8014@gmail.com> <23342.1453133686@dooku.sandelman.ca> <6ACF3DA5-AEA7-43DD-A3BA-33F680374F09@gmail.com> X-Mailer: MH-E 8.6; nmh 1.6+dev; GNU Emacs 24.4.2 X-Face: $\n1pF)h^`}$H>Hk{L"x@)JS7<%Az}5RyS@k9X%29-lHB$Ti.V>2bi.~ehC0; <'$9xN5Ub# z!G,p`nR&p7Fz@^UXIn156S8.~^@MJ*mMsD7=QFeq%AL4m Subject: Re: [Make-wifi-fast] [Cerowrt-devel] routers you can throw off the back of a truck X-BeenThere: make-wifi-fast@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Jan 2016 01:06:28 -0000 Jonathan Morton wrote: >> Jonathan Morton wrote: >>> I haven=E2=80=99t yet found a robust way to automatically sense lin= k capacity from >>> the upstream side. You=E2=80=99ll therefore need to set a conserva= tive static >>> value for the uplink capacity. >> >> As the maintainer of a PPPoE concentrator, and operator of some netw= orks, >> I've been considering whether one can estimate the bandwidth using r= ound >> trip PPP IPCP keep alives. Clearly, if both ends participate in ti= me >> stamping then it is much better, but I've been wondering if we can d= o some >> incremental deployment on one side or the other. >> >> Sadly, I mostly just think about this while cycling; I haven't writt= en any >> code yet. > In most PPPoE deployments I know about, there is also a modem from > which the actual, precise link rates can usually be queried. Where > that=E2=80=99s not the case, IPCP (or is it LPCP?) probes would be a = reasonable > workaround, but it must still be understood that the signal it provid= es > is only valid under saturating traffic, which complicates > implementation. Yes, you are rtight, I want to do LPCP echo requests. The modem might know what speed it has with the tower/DSLAM, but won't know= how congested the backhaul link is. There are some third party/white label 3G arrangements in Canada that use PPP/L2TP back to the third-party provider, but most route the IPv4 (only) packets via IPsec or MPLS. -- ] Never tell me the odds! | ipv6 mesh network= s [ ] Michael Richardson, Sandelman Software Works | network architect= [ ] mcr@sandelman.ca http://www.sandelman.ca/ | ruby on rails = [