From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mout.gmx.net (mout.gmx.net [212.227.15.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id 756C63B29E for ; Tue, 24 Jul 2018 13:58:42 -0400 (EDT) Received: from [192.168.1.133] ([79.192.253.1]) by mail.gmx.com (mrgmx002 [212.227.17.190]) with ESMTPSA (Nemesis) id 0LxgHz-1g5RYr0joh-017IHJ; Tue, 24 Jul 2018 19:58:38 +0200 Date: Tue, 24 Jul 2018 19:58:33 +0200 User-Agent: K-9 Mail for Android In-Reply-To: <09E00B24-63DA-42B3-AC60-155AC76C0CC4@darbyshire-bryant.me.uk> References: <31193A5B-88BF-4EA6-9BF6-530564B93228@darbyshire-bryant.me.uk> <09E00B24-63DA-42B3-AC60-155AC76C0CC4@darbyshire-bryant.me.uk> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----4F8IMDOIV61ZAM7FL62YP5YYC7AVOL" Content-Transfer-Encoding: 7bit To: Kevin Darbyshire-Bryant , Dave Taht CC: Cake List From: Sebastian Moeller Message-ID: <9CAB4FAB-DFCA-4F3B-8C20-ADA6B3F54F2A@gmx.de> X-Provags-ID: V03:K1:2L33oaCwBLPy4bA/9Q2NDOXrlX+d+68p5MQM9IL6+k28BU7j+YN ogeAv6voXtf7Onx0B//ssEDeou8t07Q0YkhdDep4TAnvIUloZPig5o3yloj0UrZtP9tr8Kp orcKinkCDwfNsMMimnm3HKEprOEWqr71oZ9iYEFabKSMr00C29yOF6N+/MuS95t9caYZlGA oQAkBlFRE9q+ATxJXLUMw== X-UI-Out-Filterresults: notjunk:1;V01:K0:hIcZ1eO9XdU=:GMUFxHdRBeOPjxTR3+S2lZ 9aBWCYv0zLgsBMU6hBuYWG8J4GtlFxHCLtzThyxNsyRcKJuCdqjKFlzv6rT6FNIKeHUvWiyW9 YjQY3P8EA3UwwtOYp+hgyW+DPASbdFF+x0Pq7RQLXx4GZcHGXgc0sYDtnsspn/4LdhIFLXAsL cwtOHNlBBWJfmgey+u3EZh/fotW5HViP0unzSbCPe4twqHrAieSmdp8Ig1Bg2IM+GdBxSAeQF Pk7DdMo6Zf//JRe2GkT1u6S38uYBd8Cwb+C3lxJIunLvWdLMKu/qOEdJYYbsT0xNeUkC+PBOY qpXuT2KYfok1ADlTeD0tm8/rZwqDKGvI7G49OkvjZtVUIlvjOKQCdjt7zOwTQM5kNh1Lq80CY BaMpF03gG0cDrguBIeP9t9Jbql3heDXTBuGOTQU9pPv6EyCLa8c7Cf89C7lDIMXF8ZKunJVVm aieDge0hy54tkY9ienXWwmWiahVgU53Xujj6Um/qEag0aitW91UB1T+5VZyOTTD6OLPLUDHHS 0VjmTBwqqn8Mtfgu8lyfbF+L0Uw3ZM/cA3NlxYGx9CQCoGZ9fyplYS0nczOdUWyW+X7fJiX8L yUHGZqpXv5oU5i/m/vEEXONppu327BF6Fg1EAQUEDeL44n70hiDeXqE7kj+pguX0PmyxuYWep 5cvbb4XW8lpgMXI1/CcYhPxhAvnTs8CfLi1VfFZRdEC+4fJwOrrBEKN4d6l2Uyqm1LpN8jlao +7B9kjuZVHmAfNqAlqo/nSHK4ki6fjQpLW39QvfhitxP+DlDydkpLhFpVN2T4b84Y89jRiRRl psnDuCO Subject: Re: [Cake] inbound cake or fq_codel shaping fails on cable on netflix reno X-BeenThere: cake@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: Cake - FQ_codel the next generation List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 24 Jul 2018 17:58:42 -0000 ------4F8IMDOIV61ZAM7FL62YP5YYC7AVOL Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Hi Kevin,=20 Maybe don't use the ptm keyword, but rather reduce the shaper rates to at = most 98=2E46 percent of the sync rate=2E=2E=2E On July 24, 2018 4:54:01 PM GMT+02:00, Kevin Darbyshire-Bryant wrote: > > >> On 24 Jul 2018, at 14:51, Dave Taht wrote: >>=20 >> Now that you can join the party, I note there IS a flent server in >england with irtt on it=2E >>=20 >> flent-london=2Ebufferbloat=2Enet > >Ok, well if you=E2=80=99re desperately interested=E2=80=A6=2E=2E :-) > >Plot I did running to the london server, one shaped at 99% downstream, >the other at 98%=E2=80=A6=2E and I=E2=80=99ve decided to stick at 98% as = a result - plot >is a little bit hard to distinguish but I=E2=80=99d say there=E2=80=99s a= round 8ms avg >less latency-ish between the two=2E And an order of magnitude between >using london v fremont :-) > >The upstream barely registers > ># tc -s qdisc show dev ifb4eth0 >qdisc cake 801a: root refcnt 2 bandwidth 78400Kbit diffserv3 >dual-dsthost nat ingress split-gso rtt 100=2E0ms ptm overhead 26 --=20 Sent from my Android device with K-9 Mail=2E Please excuse my brevity=2E ------4F8IMDOIV61ZAM7FL62YP5YYC7AVOL Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable Hi Kevin,

Maybe don't use the ptm keyword, but rather reduce the shaper rates to= at most 98=2E46 percent of the sync rate=2E=2E=2E

On July 24, 2018 4:54:01 PM GMT+02:00, Kevin Darbyshire-Bryant &l= t;kevin@darbyshire-bryant=2Eme=2Euk> wrote:


On 24 J= ul 2018, at 14:51, Dave Taht <dave=2Etaht@gmail=2Ecom> wrote:

Now that you can join the party, I note there IS a flent server in en= gland with irtt on it=2E

flent-london=2Ebufferbloat=2Enet

Ok, well if you=E2= =80=99re desperately interested=E2=80=A6=2E=2E :-)
Plot I did running to the london server, = one shaped at 99% downstream, the other at 98%=E2=80=A6=2E and I=E2=80=99ve= decided to stick at 98% as a result - plot is a little bit hard to disting= uish but I=E2=80=99d say there=E2=80=99s around 8ms avg less latency-ish be= tween the two=2E  And an order of magnitude between using london v fre= mont :-)

The upst= ream barely registers

# tc -s qdisc show dev ifb4eth0
qdisc cake 801a: root refcnt 2 b= andwidth 78400Kbit diffserv3 dual-dsthost nat ingress split-gso rtt 100=2E0= ms ptm overhead 26 



--
Sent from my Android device with K-9 Mail=2E Please excuse my brevity=2E ------4F8IMDOIV61ZAM7FL62YP5YYC7AVOL--