From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-la0-x22a.google.com (mail-la0-x22a.google.com [IPv6:2a00:1450:4010:c03::22a]) (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 AB28B21F2AA for ; Fri, 21 Nov 2014 03:51:35 -0800 (PST) Received: by mail-la0-f42.google.com with SMTP id s18so4129820lam.1 for ; Fri, 21 Nov 2014 03:51:32 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=BMQJ3SmzQwiUp3ZzgtWDNHUlLSSq9Vw/MwTGyMcN/XM=; b=lY2ol3GQuLEaHPJDbn2edNXM6gHRYRrk81rz1ktUQK4DZA850QurgAIvFc14JwugaS Gc6/IrGggsCN6Sg6QrNuVan1oAb4u3qJ+S5jf4GnG3+RbTpcjXLhfG+Aw+HPJG2LN/7k YjgDdhqfgrh2ipCR4BlPRSxt4eCtVFnOHiH7A0Ywo7yIjxIob7Z054xLGEnxmOlI96dM cN+kPiqiGJyYGpoOwaUEArzwVcOyr3mLprsGhd+VggOAjj4xa1+zBJjF/5S+mxc2HNkv pwv60gZ1PeZQxEq+MsflYCu0zYZoDXMwEV3APGrIPMODCxkBYOvuGO4ESPCGYJNyV4MN gtnQ== X-Received: by 10.152.7.193 with SMTP id l1mr3871235laa.57.1416570692591; Fri, 21 Nov 2014 03:51:32 -0800 (PST) MIME-Version: 1.0 Received: by 10.25.36.194 with HTTP; Fri, 21 Nov 2014 03:51:12 -0800 (PST) In-Reply-To: <0A4731BD-C9C0-4CA3-8147-C65CFF291741@gmx.de> References: <0A4731BD-C9C0-4CA3-8147-C65CFF291741@gmx.de> From: Dane Medic Date: Fri, 21 Nov 2014 12:51:12 +0100 Message-ID: To: Sebastian Moeller Content-Type: multipart/alternative; boundary=001a11c23dbcee636b05085d1016 Cc: "cerowrt-devel@lists.bufferbloat.net" Subject: Re: [Cerowrt-devel] Torrents are too fast X-BeenThere: cerowrt-devel@lists.bufferbloat.net X-Mailman-Version: 2.1.13 Precedence: list List-Id: Development issues regarding the cerowrt test router project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 21 Nov 2014 11:52:04 -0000 --001a11c23dbcee636b05085d1016 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Hi, correct, I've left egress on 4000 (I thought I should test only ingress first), but when I did the test with youtube + torrent transfer, I've had set upload to very small number (I think 10 kbytes/s). Yes I have DSL line (VDSL2 as I've read ISP info), I'll try to set the linklayer option on ethernet with overhead of 8, like it says on http://www.bufferbloat.net/projects/cerowrt/wiki/Setting_up_SQM_for_CeroWrt= _310 I have linux machine so I'll install netperf-wrapper and test things sqm configuration: root@cerowrt:~# uci show sqm sqm.ge00=3Dqueue sqm.ge00.interface=3Dge00 sqm.ge00.qdisc=3Dfq_codel sqm.ge00.script=3Dsimple.qos sqm.ge00.qdisc_advanced=3D1 sqm.ge00.ingress_ecn=3DECN sqm.ge00.egress_ecn=3DNOECN sqm.ge00.qdisc_really_really_advanced=3D1 sqm.ge00.itarget=3Dauto sqm.ge00.etarget=3Dauto sqm.ge00.linklayer=3Dnone sqm.ge00.download=3D3100 sqm.ge00.upload=3D4000 sqm.ge00.enabled=3D1 Thank you guys, Dane 2014-11-21 12:16 GMT+01:00 Sebastian Moeller : > Hi Dane, > > > On Nov 20, 2014, at 15:13 , Dane Medic wrote: > > > dpreed, thank you for response. I'm already using fq_codel with cerowrt > and I don't think it does what I want (or maybe I want too much :) > > > > So the steps I've made: > > flashed wndr3700v2 with cerowrt 3.10.50-1 then I've measured: > > > > root@cerowrt:/usr/lib/CeroWrtScripts# sh betterspeedtest.sh -p > wlan-si.net -t 120 > > 2014-11-20 12:18:34 Testing against netperf.bufferbloat.net (ipv4) with > 5 simultaneous sessions while pinging wlan-si.net (120 seconds in each > direction) > > > .........................................................................= ................................................ > > Download: 3.78 Mbps > > Latency: (in msec, 119 pings, 0.00% packet loss) > > Min: 13.077 > > 10pct: 251.522 > > Median: 317.851 > > Avg: 308.497 > > 90pct: 371.033 > > Max: 376.132 > > > .........................................................................= ................................................... > > Upload: 0.48 Mbps > > Latency: (in msec, 103 pings, 0.00% packet loss) > > Min: 12.278 > > 10pct: 12.727 > > Median: 18.359 > > Avg: 23.256 > > 90pct: 33.971 > > Max: 180.303 > > > > Then I've put these commands: > > > > uci set sqm.ge00.enabled=3D1 > > uci set sqm.ge00.download=3D3200 > > uci set sqm.ge00.qdisc=3Dnfq_codel > > uci commit sqm > > reboot > > > > And another measure: > > > > root@cerowrt:/usr/lib/CeroWrtScripts# sh betterspeedtest.sh -p > wlan-si.net -t 120 > > 2014-11-20 12:49:05 Testing against netperf.bufferbloat.net (ipv4) with > 5 simultaneous sessions while pinging wlan-si.net (120 seconds in each > direction) > > > .........................................................................= ................................................ > > Download: 2.74 Mbps > > Latency: (in msec, 121 pings, 0.00% packet loss) > > Min: 12.210 > > 10pct: 13.002 > > Median: 15.077 > > Avg: 15.095 > > 90pct: 16.968 > > Max: 18.599 > > > .........................................................................= .................................................... > > Upload: 0.49 Mbps > > Latency: (in msec, 101 pings, 0.00% packet loss) > > Min: 12.255 > > 10pct: 12.684 > > Median: 16.679 > > Avg: 23.100 > > 90pct: 34.019 > > Max: 170.173 > > > > The tests doesn't look bad, but the problem is I watch a video clip on > youtube and my sister starts torrent client, I can't watch anymore. > > Could you post the content of /etc/config/sqm from after > activating SQM please. It looks like you did not activate shaping on egre= ss > (its not in the =E2=80=9Cici set *=E2=80=9D above and the upload statisti= cs look identical > to the first unshaped example). You really need to control the buffer in > both directions to get rid of nasty latency spikes (especially with > torrents that I assume will try to use both directions maximally if left > to their own devices). Also 4M/512K sounds like a DSL link, if so you mig= ht > find the link layer adjustments helpful (if there are question what to fi= ll > in just ask). Also if you have a linux or macosx computer available I wou= ld > recommend to install netperf-wrapper and use the RRUL test to > simultaneously load down- and up-link (or alternatively netperfrunner.sh) > as this will show bloated buffers more clearly than the individual tests > for each direction as performed by betterspeedtest.sh. > > Best Regards > Sebastian > > > > > > Cheers > > _______________________________________________ > > Cerowrt-devel mailing list > > Cerowrt-devel@lists.bufferbloat.net > > https://lists.bufferbloat.net/listinfo/cerowrt-devel > > --001a11c23dbcee636b05085d1016 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
Hi,

correct, I've left egr= ess on 4000 (I thought I should test only ingress first), but when I did th= e test with youtube + torrent transfer, I've had set upload to very sma= ll number (I think 10 kbytes/s).

Yes I have DSL line (VDSL2 as= I've read ISP info), I'll try to set the linklayer option on ether= net with overhead of 8, like it says on http://www.bufferb= loat.net/projects/cerowrt/wiki/Setting_up_SQM_for_CeroWrt_310
= I have linux machine so I'll install netperf-wrapper and test things

sqm configuration:

root@cerowrt:~# uci show sqm
sq= m.ge00=3Dqueue
sqm.ge00.interface=3Dge00
sqm.ge00.qdisc=3Dfq_codelsqm.ge00.script=3Dsimple.qos
sqm.ge00.qdisc_advanced=3D1
sqm.ge00.in= gress_ecn=3DECN
sqm.ge00.egress_ecn=3DNOECN
sqm.ge00.qdisc_really_rea= lly_advanced=3D1
sqm.ge00.itarget=3Dauto
sqm.ge00.etarget=3Dauto
s= qm.ge00.linklayer=3Dnone
sqm.ge00.download=3D3100
sqm.ge00.upload=3D4= 000
sqm.ge00.enabled=3D1

Thank you guys,
Dane

2014-11-21 12:16 GMT+01:00 Sebastian Moeller &l= t;moeller0@gmx.de&= gt;:
Hi Dane,


On Nov 20, 2014, at 15:13 , Dane Medic <dm70dm@gmail.com> wrote:

> dpreed, thank you for response. I'm already using fq_codel with ce= rowrt and I don't think it does what I want (or maybe I want too much := )
>
> So the steps I've made:
> flashed wndr3700v2 with cerowrt 3.10.50-1 then I've measured:
>
> root@cerowrt:/usr/lib/CeroWrtScripts# sh betterspeedtest.sh -p wlan-si.net -t 120
> 2014-11-20 12:18:34 Testing against netperf.bufferbloat.net (ipv4) with 5 simult= aneous sessions while pinging wlan-si.net (120 seconds in each direction)
> ......................................................................= ...................................................
>=C2=A0 Download:=C2=A0 3.78 Mbps
>=C2=A0 =C2=A0Latency: (in msec, 119 pings, 0.00% packet loss)
>=C2=A0 =C2=A0 =C2=A0 =C2=A0Min: 13.077
>=C2=A0 =C2=A0 =C2=A010pct: 251.522
>=C2=A0 =C2=A0 Median: 317.851
>=C2=A0 =C2=A0 =C2=A0 =C2=A0Avg: 308.497
>=C2=A0 =C2=A0 =C2=A090pct: 371.033
>=C2=A0 =C2=A0 =C2=A0 =C2=A0Max: 376.132
> ......................................................................= ......................................................
>=C2=A0 =C2=A0 Upload:=C2=A0 0.48 Mbps
>=C2=A0 =C2=A0Latency: (in msec, 103 pings, 0.00% packet loss)
>=C2=A0 =C2=A0 =C2=A0 =C2=A0Min: 12.278
>=C2=A0 =C2=A0 =C2=A010pct: 12.727
>=C2=A0 =C2=A0 Median: 18.359
>=C2=A0 =C2=A0 =C2=A0 =C2=A0Avg: 23.256
>=C2=A0 =C2=A0 =C2=A090pct: 33.971
>=C2=A0 =C2=A0 =C2=A0 =C2=A0Max: 180.303
>
> Then I've put these commands:
>
> uci set sqm.ge00.enabled=3D1
> uci set sqm.ge00.download=3D3200
> uci set sqm.ge00.qdisc=3Dnfq_codel
> uci commit sqm
> reboot
>
> And another measure:
>
> root@cerowrt:/usr/lib/CeroWrtScripts# sh betterspeedtest.sh -p wlan-si.net -t 120
> 2014-11-20 12:49:05 Testing against netperf.bufferbloat.net (ipv4) with 5 simult= aneous sessions while pinging wlan-si.net (120 seconds in each direction)
> ......................................................................= ...................................................
>=C2=A0 Download:=C2=A0 2.74 Mbps
>=C2=A0 =C2=A0Latency: (in msec, 121 pings, 0.00% packet loss)
>=C2=A0 =C2=A0 =C2=A0 =C2=A0Min: 12.210
>=C2=A0 =C2=A0 =C2=A010pct: 13.002
>=C2=A0 =C2=A0 Median: 15.077
>=C2=A0 =C2=A0 =C2=A0 =C2=A0Avg: 15.095
>=C2=A0 =C2=A0 =C2=A090pct: 16.968
>=C2=A0 =C2=A0 =C2=A0 =C2=A0Max: 18.599
> ......................................................................= .......................................................
>=C2=A0 =C2=A0 Upload:=C2=A0 0.49 Mbps
>=C2=A0 =C2=A0Latency: (in msec, 101 pings, 0.00% packet loss)
>=C2=A0 =C2=A0 =C2=A0 =C2=A0Min: 12.255
>=C2=A0 =C2=A0 =C2=A010pct: 12.684
>=C2=A0 =C2=A0 Median: 16.679
>=C2=A0 =C2=A0 =C2=A0 =C2=A0Avg: 23.100
>=C2=A0 =C2=A0 =C2=A090pct: 34.019
>=C2=A0 =C2=A0 =C2=A0 =C2=A0Max: 170.173
>
> The tests doesn't look bad, but the problem is I watch a video cli= p on youtube and my sister starts torrent client, I can't watch anymore= .

=C2=A0 =C2=A0 =C2=A0 =C2=A0 Could you post the content of /etc/= config/sqm from after activating SQM please. It looks like you did not acti= vate shaping on egress (its not in the =E2=80=9Cici set *=E2=80=9D above an= d the upload statistics look identical to the first unshaped example). You = really need to control the buffer in both directions to get rid of nasty la= tency spikes (especially with torrents that I assume=C2=A0 will try to use = both directions maximally if left to their own devices). Also 4M/512K sound= s like a DSL link, if so you might find the link layer adjustments helpful = (if there are question what to fill in just ask). Also if you have a linux = or macosx computer available I would recommend to install netperf-wrapper a= nd use the RRUL test to simultaneously load down- and up-link (or alternati= vely netperfrunner.sh) as this will show bloated buffers more clearly than = the individual tests for each direction as performed by betterspeedtest.sh.=

Best Regards
=C2=A0 =C2=A0 =C2=A0 =C2=A0 = Sebastian


>
> Cheers
> _______________________________________________
> Cerowrt-devel mailing list
> Cerowrt-devel@l= ists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cerowrt-devel


--001a11c23dbcee636b05085d1016--