From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ee0-f43.google.com (mail-ee0-f43.google.com [74.125.83.43]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority" (verified OK)) by huchra.bufferbloat.net (Postfix) with ESMTPS id 908BA21F0EA for ; Fri, 26 Oct 2012 05:39:34 -0700 (PDT) Received: by mail-ee0-f43.google.com with SMTP id c13so1268011eek.16 for ; Fri, 26 Oct 2012 05:39:32 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; bh=3l7k6dB1t/vdzqp79t5g38DjSkr1IhXPTteeFDuT3uo=; b=s4Ys1KI78rDeICnBUEDVg5rmkqBzyMB0jvDx2662IUaxDMAX7pM86JnEGW5cA8Ia4U 1SzQyFk1W+t7kpDut9TA+bpGfpg1gdf0IldbaetPiBA3BkBqI/aN6ZRn+YYFmsEl0aQl LZGSKJuJFVeu0JS0A+W6FpLHME1NZC4JVXRX3+E/6JSJuQkeCNOpd0KOEpxlUl/J1WXF iNalQ+drrKCBlj1JOYTJtkcCrFqLtbF2OdSTMct0tjaRDs02KIX7o+Wr+nXKpv7T9Sk/ +AQZR8wUbQosTYLyUd7zi5Vl9qPT0+0APkCWd+tvnQPoxtYnCMbmVG9IiAzSZI8enssS HJSw== MIME-Version: 1.0 Received: by 10.14.200.134 with SMTP id z6mr31870910een.33.1351255172614; Fri, 26 Oct 2012 05:39:32 -0700 (PDT) Received: by 10.223.180.10 with HTTP; Fri, 26 Oct 2012 05:39:32 -0700 (PDT) In-Reply-To: <20121026065558.GA3117@merlins.org> References: <20121026042902.GG8019@merlins.org> <20121026065558.GA3117@merlins.org> Date: Fri, 26 Oct 2012 08:39:32 -0400 Message-ID: From: Dave Taht To: Marc MERLIN Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Cc: cerowrt-users@lists.bufferbloat.net Subject: Re: [Cerowrt-users] Some setup questions X-BeenThere: cerowrt-users@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: Support for user problems regarding cerowrt List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 26 Oct 2012 12:39:35 -0000 On Fri, Oct 26, 2012 at 2:55 AM, Marc MERLIN wrote: > On Thu, Oct 25, 2012 at 09:29:02PM -0700, Marc MERLIN wrote: >> 2) I'm going to need usb to serial (pl2303) support. I already read that= cerowrt >> doesn't have most packages from openwrt, but since they should be mostly= compatible >> can't I just point cerowrt to >> http://downloads.openwrt.org/attitude_adjustment/12.09-beta2/ar71xx/gene= ric/packages/Packages >> and use those packages instead, including >> http://downloads.openwrt.org/attitude_adjustment/12.09-beta2/ar71xx/gene= ric/packages/kmod-usb-serial-pl2303_3.3.8-1_ar71xx.ipk ? > > Actually as a followup, considering that I need to use my WNDR3800 as a r= ock > solid internet gateway for my house, and it will double as a serial conso= le > server for my server (hence the need for 3rd party modules and packages t= hat > are in the openwrt tree). As noted the serial modules are > I won't be reflashing it often, ideally once every few years at most, and= on > top of wanting to benefit the fq_codel work (I already signed up for a > monetary donation to help pay for it), all I see myself using is: I have appreciated all the donations since I started soliciting for them a month ago. It made the difference between living under a bridge, and living in a yurt, that month. http://www.bufferbloat.net/projects/cerowrt/wiki/yurtlab The outpouring of support was wonderful and encouraging and I'd have never got sugarland out the door if that hadn't happened. That said, the donations/subscriptions so far are not suffiicient presently to carry the project very far. As one example, I'd like to repeat the field test above in the april timeframe, with some fixed wireless code. I've spent a lot of time these past two months trying to find ways to raise enough money to fund another 18 month development cycle, solving the as yet unsolved problems with wireless, working on mdns issue, dlna, and the like - but getting a non-profit off the ground is going to take time, and running a kickstarter/indigogo campaign, while helpful, seems unlikely to generate enough funds to really go after those problems. I'm (teklibre) prepared to fund at least part of the next development cycle, after I get unburied from my new contract and the mountain of debt accumulated in support of bufferbloat.net reduced somewhat. The research benefit from the bufferbloat project has been enormous - not just the development of fq_codel, but the insights gained from a hard look at the device driver problems (BQL, TCP small queues, codel), from routing everything, trying to make ipv6 work, tracking the work of the homenet and mdnsext working groups, etc, so I would hope to find matching funds from a set of corporate donors for the next round. And I'm tickled that so many notables (dave miller, eric dumazet, david reed, esr, jg, dozens of others) are running cerowrt on a day to day basis. And delighted by all the hard work, development, and testing, by the openwrt folk and the cerowrt volunteers in getting to this point in 18 months... and the work in the mainline kernel... and the theoretical work by dozens of academics and then of course there's the huge contributions of van, kathie, edumazet, and tom herbert in fixing linux, felix on the ath9k... but I'm pretty tapped out personally through the end of the year. There is some good work going on that will make the openwrt and cerowrt development process easier - in particular a 4 node heavy duty cluster is going into place to help with continuous integration, thx to the hard work of denis and john, openwrt is streamlining their processes, and I'm trying to spread the knowlege of how to build the darn thing around. I'm heading up to linaro's event in the EU this weekend, and linuxcon the next, in the hope that I can convince some org/co to help out more on the money side. I seem to have missed your contribution on this page: http://cero2.bufferbloat.net/cerowrt/credits.html I will go look at the backlog from paypal. > - bind I ripped bind out of the sugarland build as we'd encountered memory use problems on 64MB and lower systems and went back to dnsmasq. Also bind was on it's 5th CVE for the year. I was kind of fed up. It's still easily installable (and will work fine on a 3800), and I like to think it's the most robust implementation of bind ever done - and somewhere on this I documented how to install bind-latest and bind-latest-config on top of sugarland, or I put it on the wiki... > - dhcp In this case we are using dnsmasq for dhcp > - minicom > - screen I can add those to the next build. Probably. > This is what I currently have on my old WRT-600N with dd-wrt, but dd-wrt = is > obsolete, and the router's wireless is poor, hence my upgrading. > > Given that, am I better off with openwrt 12.09-beta2 if I want a long ter= m > stable build or the latest cerowrt? I would wait on both counts - wait for openwrt to go final, wait for cerowrt to sync up. And in the meantime, play a little with both sets of ideas. A huge core difference between openwrt and cerowrt is that openwrt is bridged, cerowrt is routed. Cerowrt also does a few other things differently, and comes with a boatload of test tools, mostly documented here: http://cero2.bufferbloat.net/cerowrt/about.html However, given your desire for multi-year LTS, that really isn't in the cards for cerowrt. It is my hope that we'll get a cerowrt-3.3.8 stable release soon, but i wasn't planning on maintaining that for more than a few months, preferring to launch back into tracking 3.6 and later with new de-bufferbloating fixes that are waiting in the wings. I > > Thanks, > Marc > -- > "A mouse is a device used to point at the xterm you want to type in" - A.= S.R. > Microsoft is to operating systems .... > .... what McDonalds is to gourmet c= ooking > Home page: http://marc.merlins.org/ > _______________________________________________ > Cerowrt-users mailing list > Cerowrt-users@lists.bufferbloat.net > https://lists.bufferbloat.net/listinfo/cerowrt-users --=20 Dave T=E4ht Fixing bufferbloat with cerowrt: http://www.teklibre.com/cerowrt/subscribe.= html