From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mout.gmx.net (mout.gmx.net [212.227.17.20]) (using TLSv1 with cipher DHE-RSA-AES128-SHA (128/128 bits)) (Client CN "mout.gmx.net", Issuer "TeleSec ServerPass DE-1" (verified OK)) by huchra.bufferbloat.net (Postfix) with ESMTPS id 863F821F112 for ; Sat, 21 Dec 2013 02:32:37 -0800 (PST) Received: from [10.17.151.121] ([80.187.96.24]) by mail.gmx.com (mrgmx002) with ESMTPSA (Nemesis) id 0LxPAo-1VSJQm1aQc-016zbB for ; Sat, 21 Dec 2013 11:32:34 +0100 User-Agent: K-9 Mail for Android In-Reply-To: References: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable From: Sebastian Moeller Date: Sat, 21 Dec 2013 11:32:24 +0100 To: Hector Ordorica , cerowrt-devel@lists.bufferbloat.net Message-ID: <1b73abc0-ba7e-4b87-b786-0df6562f8a9b@email.android.com> X-Provags-ID: V03:K0:s9h8HPIQNmpm8qgIEZT9QuYsv0vZwKh4WiRECMihji2UA9Bf/z2 gIrD9wmYwFxsEwhvycVQ8GdNqUUARSsmEo9d//YRnN4GIsRivXS+cQ/iecWjnTvC5UBA+9T REFZ4YP4vpS3ixWLgSCRqZgZO+uaKPVis0uRljiXR0hLWWpp8/KbVT2YG1ROetPYUJTH4mE zS/MPDbL8stQ4Fwu5bZTQ== Subject: Re: [Cerowrt-devel] Proper AQM settings for my connection? 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: Sat, 21 Dec 2013 10:32:38 -0000 Hector Ordorica wrote: >I'm running 3=2E10=2E13-2 on a WNDR3800, and have used the suggested >settings from the latest draft: > >http://www=2Ebufferbloat=2Enet/projects/cerowrt/wiki/Setting_up_AQM_for_C= eroWrt_310 > >I have a 30Mb down / 5Mb upload cable connection=2E > >With fq_codel, even undershooting network upload bandwidth by more >than 95%, I'm seeing 500ms excessive upload buffering warnings from >netalyzr=2E Download is ok at 130ms=2E I was previously on a 3=2E8 releas= e >and the same was true=2E So I have been fooled by netalyzr before just as you now=2E Netalyzr uses = a very peculiar probe to measure the depth of the buffers: a totally nonrea= ctive inelastic "flood" of UDP packets of relative short duration=2E The on= ly real world traffic that looks like this is a denial of service attack on= your router=2E Fq_codel tries very hard to be a good citizen that steers f= lows gently to their fair share of the bandwidth, in case flows do not reac= t fq_codel will slowly take the gloves of so to say and restrict these flow= s more aggressively=2E The netalyzr probe now is too short for fq_codel to = actually get serious in its packet dropping=2E Now real traffic typically, = be it TCP or UDP tries to adjust to dropped packets by reducing the transmi= ssion rate=2E In other words netalyzt measures a sort of worst case bufferi= ng for fq_codel=2E Note for pfifo_fast this worst case is actually somethin= g you encounter with real traffic as well=2E So what netalyzr is missing is= a report telling you whether the reported buffering will increase the over= all latency of the system, or not=2E=2E=2E=2E To summarize unless you see UDP floods as a typical use case for you= r internet connection, the netalyzr buffering numbers have no great signifi= cance for day to day use of your internet connection, if your are using a m= odern qdisc like fq_codel or pie=2E As Dave taught me in the past, you can easily test this hypothesis by= modifying the limit parameter of fq_codel in simple, =2Eqos or simplest=2Eqos=2E The larger limit and the slower the link speed= in the measured direction the greater the reported buffering=2E > >With pie (and default settings), the buffer warnings go away: > >http://n2=2Enetalyzr=2Eicsi=2Eberkeley=2Eedu/summary/id=3D43ca208a-32182-= 9424fd6e-5c5f-42d7-a9ea > >And the connection performs very well while torrenting and gaming=2E > >Should I try new code? Or can I tweak some variables and/or delay >options in scripts for codel? > >Thanks for your work, >Hector >_______________________________________________ >Cerowrt-devel mailing list >Cerowrt-devel@lists=2Ebufferbloat=2Enet >https://lists=2Ebufferbloat=2Enet/listinfo/cerowrt-devel Hi Hector, --=20 Sent from my Android phone with K-9 Mail=2E Please excuse my brevity=2E