From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-qk0-x234.google.com (mail-qk0-x234.google.com [IPv6:2607:f8b0:400d:c09::234]) (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 5147A21F868 for ; Wed, 30 Sep 2015 11:14:03 -0700 (PDT) Received: by qkas79 with SMTP id s79so21436126qka.0 for ; Wed, 30 Sep 2015 11:14:02 -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; bh=XaID1MAVIzgUNZ3Nmkb6Y9D8mgoEeGTkBYrfZZCsaQY=; b=i1AQdAc0xLUpLGLMiVwvc5f9pIYK/lWe7HzSljD6Xt2zIGxUcxreUH4Izz6jwfdARQ Up3Jtgv1j65//oL+RD5d0F7Xrqr7s1kNgd9tmZkIDZCGTh5EwPKXhZgpwE82gUUSccXl MNVVOJVOYlVtLSC8uOGGazkXi/70NhfhAUR1b4iubszJG2rMltnaWvt5VrIFISly/ZT0 STHXD16JMZNWli/XwTDHVPRpClZV4O+QO9CwODGvyh/OEOQWNpgoowLbiBvlLhyogU4H jEiLSBHLaNFPDwILKrc4Qu7pS2VGpG6NcUTdbq0SoVMdyPb1M4fSJ61cS/ZdJLlu7H5S RW+Q== MIME-Version: 1.0 X-Received: by 10.55.209.68 with SMTP id s65mr6100344qki.97.1443636842830; Wed, 30 Sep 2015 11:14:02 -0700 (PDT) Received: by 10.140.85.20 with HTTP; Wed, 30 Sep 2015 11:14:02 -0700 (PDT) In-Reply-To: References: Date: Wed, 30 Sep 2015 19:14:02 +0100 Message-ID: From: Alan Jenkins To: Kevin Darbyshire-Bryant Content-Type: text/plain; charset=UTF-8 Cc: "cake@lists.bufferbloat.net" Subject: Re: [Cake] Cubic issues? 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: Wed, 30 Sep 2015 18:14:26 -0000 On 29/09/2015, Kevin Darbyshire-Bryant wrote: > This is interesting > http://www.theregister.co.uk/2015/09/29/google_cubists_fix_congestion_control_for_faster_tcp/ heh, I would recommend reading their source because they seemed to omit the point http://bitsup.blogspot.co.uk/2015/09/thanks-google-tcp-team-for-open-source.html [Patrick McManus] The problem can be roughly summarized as the controller mistakenly characterizing the lack of congestion reports *over a quiescent period* as positive evidence that the network is not congested [thereg] the bug was simple: TCP CUBIC interprets a lack of congestion reports as an opportunity to send data at a faster rate. [thereg] That condition could, however, arise merely because the system hasn't been getting any congestion reports. i.e. someone choked on the word quiescent and forgot to re-word it, leaving a tautology. Alan