From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-qg0-x232.google.com (mail-qg0-x232.google.com [IPv6:2607:f8b0:400d:c04::232]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (Client CN "smtp.gmail.com", Issuer "Google Internet Authority G2" (verified OK)) by huchra.bufferbloat.net (Postfix) with ESMTPS id 3D48A21F963 for ; Sat, 25 Jul 2015 13:58:19 -0700 (PDT) Received: by qged69 with SMTP id d69so29359775qge.0 for ; Sat, 25 Jul 2015 13:58:18 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=date:mime-version:message-id:in-reply-to:references:from:to:subject :content-type; bh=OynotwpDlUbUq8mxvul8r57sXPeNVvePSN5ZtSbtK64=; b=eO62JdZgpAfM8XCCzbcJ469LzESve7qNScxWwiZmppnfShQrTzIN+g+SeIu35/2Ym0 NlswrfeB+gy/jlPH5c00tSQVqwjIHTy2dleItFQvgoeCy10t5RuuyyGWMFpizH3nqRMS x7hGrpHpRQjrmlWc0uaJqO+9yInXXXaTyFeIYbsYYfgyO8UtQUghoCaDnkuMyvob8fWJ l85vwJAIX8Z62+MoVD6PqrAx6aA72LNiawiLH0woCnzJ0T0l60f8fHJdTXJgagiQJ78Q KvVPjoS37K+sA7UmHkRxXdPtcM9TCF9polYq92wyf3B6zAk57dIJKS6i/yxNowOI4eJs PGFA== X-Received: by 10.140.148.204 with SMTP id 195mr16524615qhu.24.1437857898434; Sat, 25 Jul 2015 13:58:18 -0700 (PDT) Received: from hedwig-63.prd.orcali.com (ec2-54-85-253-19.compute-1.amazonaws.com. [54.85.253.19]) by smtp.gmail.com with ESMTPSA id a8sm6231700qka.0.2015.07.25.13.58.17 for (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Sat, 25 Jul 2015 13:58:17 -0700 (PDT) Date: Sat, 25 Jul 2015 13:58:17 -0700 (PDT) X-Google-Original-Date: Sat, 25 Jul 2015 20:58:17 GMT MIME-Version: 1.0 X-Mailer: Nodemailer (0.5.0; +http://www.nodemailer.com/) Message-Id: <1437857897145.8e4f82eb@Nodemailer> In-Reply-To: <2BE70C89-86F7-43F8-8AFE-FFA524189408@gmx.de> References: <2BE70C89-86F7-43F8-8AFE-FFA524189408@gmx.de> X-Orchestra-Oid: 9B0AB916-21E6-4FC4-8430-A66BC359DEF0 X-Orchestra-Sig: 3e316d926ea34f10a9406557090e24d827f07a2e X-Orchestra-Thrid: 16C35577-C0DD-4AD9-A7DC-276830CC8CF3 X-Orchestra-Thrid-Sig: 54b4d22fd1923c4c99ad00b208a14873b3b9f527 X-Orchestra-Account: 2859bc6411076f32cf7f3a708bb9672bb7b2e184 From: "Alec Robertson" To: cake@lists.bufferbloat.net Content-Type: multipart/alternative; boundary="----Nodemailer-0.5.0-?=_1-1437857897695" Subject: Re: [Cake] How to test Cake on TP-Link WDR3600 X-BeenThere: cake@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: Cake - FQ_codel the next generation List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 25 Jul 2015 20:58:48 -0000 ------Nodemailer-0.5.0-?=_1-1437857897695 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable I=E2=80=99m assuming I put =22tc -d qdisc=E2=80=9D into BusyBox=3F Quick sub-question (off-topic so my apologies), this firmware I=E2=80=99m = using that I linked to previously, has HTTPS enabled which means every time= I go to Luci I get a security error in Chrome. How do I disable HTTPS=3F -- Alec Robertson On Sat, Jul 25, 2015 at 8:20 PM, Sebastian Moeller wrote: > Hi Alec, > On Jul 25, 2015, at 20:57 , Alec Robertson = wrote: >> That's it, it's all showing up now. >>=20 >> I will try it out right away. What should I be testing and looking out = for=3F > 1) test whether using cake as link layer adaptation mechanism works: put = in a number for overhead and run =E2=80=9Ctc -d qdisc=E2=80=9D and post the= output of: > a) cat /etc/config/sqm > b) /etc/init.d/sqm stop ; tc -d qdisc > c) /etc/init.d/sqm start ; tc -d qdisc > 2) test whether you like the performance under load, http://www.= dslreports.com/speedtest is a good start, registered (for free) users can = change the configuration, I like to use 16 streams for upstream and = downstream each and I like the =E2=80=9CHi-Res BufferBloat: (10hz = check)=E2=80=9D option. > 3) test even more, have a loo at flent (https://flent.org), especially = the rrul test has been a real help in understanding link behavior under = load. > Please note, piece=5Fof=5Fcake will only use 1 priority for all packets = (so effectively no priority), layer=5Fcake.qos will by default, use 4 = different priority layers, so expect some difference between layer=5Fcake = and piece=5Fof=5Fcake in the rrul results... > Best Regards > Sebastian >> =5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F= =5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F >> Cake mailing list >> Cake@lists.bufferbloat.net >> https://lists.bufferbloat.net/listinfo/cake ------Nodemailer-0.5.0-?=_1-1437857897695 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable I=E2=80=99m assuming I put =22tc -d = qdisc=E2=80=9D into BusyBox=3F

Quick sub-question (off-topic so my apologies), this firmware = I=E2=80=99m using that I linked to previously, has HTTPS enabled which = means every time I go to Luci I get a security error in Chrome. How do I = disable HTTPS=3F

--
Alec Robertson


On Sat, Jul 25, 2015 at 8:20 PM= , Sebastian Moeller <moeller0@gmx.de> = wrote:

Hi Alec,


On Jul 25, 2015, at 20:57 , Alec Robertson = <alecrobertson13@gmail.com> wrote:

> That's it, it's all showing up now.
>=20
> I will try it out right away. What should I be testing and looking= out for=3F

1) test whether using cake as link layer adaptation mechanism = works: put in a number for overhead and run =E2=80=9Ctc -d qdisc=E2=80=9D = and post the output of:
a) cat /etc/config/sqm
b) /etc/init.d/sqm stop ; tc -d qdisc
c) /etc/init.d/sqm start ; tc -d qdisc

2) test whether you like the performance under load, http://www.= dslreports.com/speedtest is a good start, registered (for free) users can = change the configuration, I like to use 16 streams for upstream and = downstream each and I like the =E2=80=9CHi-Res BufferBloat: (10hz = check)=E2=80=9D option.

3) test even more, have a loo at flent (https://flent.org), = especially the rrul test has been a real help in understanding link = behavior under load.
Please note, piece=5Fof=5Fcake will only use 1 priority for all packets= (so effectively no priority), layer=5Fcake.qos will by default, use 4 = different priority layers, so expect some difference between layer=5Fcake = and piece=5Fof=5Fcake in the rrul results...


Best Regards
Sebastian


> =5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F= =5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F=5F= =5F=5F=5F
> Cake mailing list
> Cake@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cake


------Nodemailer-0.5.0-?=_1-1437857897695--