Re: [v6ops] ULA and IPv4 - draft-liu-v6ops-ula-usage-analysis

Mark ZZZ Smith <> Tue, 05 November 2013 20:27 UTC

Return-Path: <>
Received: from localhost (localhost []) by (Postfix) with ESMTP id A15DF11E813A for <>; Tue, 5 Nov 2013 12:27:00 -0800 (PST)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -1.906
X-Spam-Status: No, score=-1.906 tagged_above=-999 required=5 tests=[AWL=0.193, BAYES_00=-2.599, FROM_LOCAL_NOVOWEL=0.5]
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id nZLm69LTZrSP for <>; Tue, 5 Nov 2013 12:26:52 -0800 (PST)
Received: from ( []) by (Postfix) with ESMTP id B428F11E81C3 for <>; Tue, 5 Nov 2013 12:26:48 -0800 (PST)
Received: from [] by with NNFMP; 05 Nov 2013 20:26:47 -0000
Received: from [] by with NNFMP; 05 Nov 2013 20:26:47 -0000
Received: from [] by with NNFMP; 05 Nov 2013 20:26:47 -0000
X-Yahoo-Newman-Property: ymail-3
Received: (qmail 49612 invoked by uid 60001); 5 Nov 2013 20:26:47 -0000
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;; s=s1024; t=1383683207; bh=+cuoI5WwI+lvF4wakq0YjLew90nRwDixPqcz/CxhyTY=; h=X-YMail-OSG:Received:X-Rocket-MIMEInfo:X-Mailer:References:Message-ID:Date:From:Reply-To:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding; b=6vnu2ZGUuIJS2+fRenCR8mH6hqoc3Bgo733oVMQRku+zgY9jEC0zNxV5vdeZcViBC46ZK53bB0neiUq8bL4iycTfKHmfN5c0urPPxT1CrUd8dHDfpRZWwhZP1qgvz1gQsVnASPklDMZ9k8RpNSA0ZgdwfSQGq3HAueum+H6myec=
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024;; h=X-YMail-OSG:Received:X-Rocket-MIMEInfo:X-Mailer:References:Message-ID:Date:From:Reply-To:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding; b=NT+g5CXLHl14kD+ONu7yXnIDDtNc3DGkm8K/dNXxEhUsQIkqh6pTQiT6FozQKo87taFcQuH8S0rHMaRAFCPc8zgHJqKZJiGxeqcg2XTSMJ+B0XHF6uQ0ujCG+YRdIXCiqtXIGfU8IMkSj8LTv/zFjHXm5x8sbrrV19ozGwfQbaE=;
X-YMail-OSG: 0gpv5QsVM1mvSVJXiw_odNHtWMDu41S6HbTApTZMMcDVLIE RuosFRNyONWAOqKJUOgxhgdt1WELLliSesM6JuvCuEtphwvqvCTI34jB97RQ 24m5T0LeTvNguB.8EWu385xrOeDM5e09M8mDVjLQl7dnUd467LkktUCgRwuu ZqSydvWGtltfcepBVMyKThK8eBnv0MbxIdX4lrWOwJhL2WOgu4nIvehBs2ZF xqzEeKhvKTxIEqkSKLEJhCDXlf_s0s_9cyZmfULIUKaq.Q6nu9LONoQujbTO 1iTbqxpL26nekoaWB.e4m0TJoybQyvP_NXigLN6VzCyXlZt.vEl0h5.eWght jbPzJYuE9HCSdcGwi8NuePF2n5CMiPhy8hIo.qUCABJtgy8h8X3a1pqqWnEc 8L1MxXxhr86Fmh7aY13o_NF0XGn.6XP7POZSwrwMwJCCi3T6gVNNUZ5MCg2u UIt8yar3.A9J8OGKBs.7LHVgSvaxMk84S3X54rf9siqpO7XG9DHdXVsAigGw VyfPqut8XBUHwA4hNhMSssv0GFIR.dNnzTTFg86.QawY4QjUrYpNC0hpw.Rv psuyULThm91.qoxA_.nr9ETBIYvLshtqfSvh9SomcjKEVjg--
Received: from [] by via HTTP; Tue, 05 Nov 2013 12:26:47 PST
X-Rocket-MIMEInfo: 002.001, CgoKCgo.X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18KPiBGcm9tOiBMb3JlbnpvIENvbGl0dGkgPGxvcmVuem9AZ29vZ2xlLmNvbT4KPlRvOiBNYXJrIFpaWiBTbWl0aCA8bWFya3p6enNtaXRoQHlhaG9vLmNvbS5hdT4gCj5DYzogQ2hyaXN0b3BoZXIgUGFsbWVyIDxDaHJpc3RvcGhlci5QYWxtZXJAbWljcm9zb2Z0LmNvbT47IFRpbSBDaG93biA8dGpjQGVjcy5zb3Rvbi5hYy51az47ICJ2Nm9wc0BpZXRmLm9yZyIgPHY2b3BzQGlldGYub3JnPiAKPlNlbnQ6IFdlZG5lc2RheSwgNiBOb3ZlbWJlciABMAEBAQE-
X-Mailer: YahooMailWebService/
References: <> <> <EMEW3|c7700e679335ec63fa8cc5ca34b52656pA42wx03tjc||> <> <> <>
Message-ID: <>
Date: Tue, 5 Nov 2013 12:26:47 -0800 (PST)
From: Mark ZZZ Smith <>
To: Lorenzo Colitti <>
In-Reply-To: <>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Cc: "" <>
Subject: Re: [v6ops] ULA and IPv4 - draft-liu-v6ops-ula-usage-analysis
X-Mailman-Version: 2.1.12
Precedence: list
Reply-To: Mark ZZZ Smith <>
List-Id: v6ops discussion list <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Tue, 05 Nov 2013 20:27:00 -0000

> From: Lorenzo Colitti <>
>To: Mark ZZZ Smith <> 
>Cc: Christopher Palmer <>om>; Tim Chown <>uk>; "" <> 
>Sent: Wednesday, 6 November 2013 6:37 AM
>Subject: Re: [v6ops] ULA and IPv4 - draft-liu-v6ops-ula-usage-analysis
>On Wed, Nov 6, 2013 at 4:32 AM, Mark ZZZ Smith <> wrote:
>Happier Eyeballs
>Great. We can't be bothered to get things right in the network, so we just tell hosts to try everything all the time in the hope that something will work.

Networks can lose packets, regardless of whether they're IPv4 or IPv6.

I don't think the problem happy eyeballs is solving isn't really any different to the problem where multiple IPv4 A records are returned and a client has to walk through them until it succeeds. Fortunately that seems to be reliable enough that end users don't have to wait for connection timeouts as the returned list of A records is walked through. But if they were (are?) waiting, then happy-eyeballs for only IPv4 would also be beneficial.

The happy eyeballs technique could be considered to be another form of retransmission to recover from packet loss. A host has to recover from packet loss whether it is within a single protocol, or across multiple ones used by and for the same application.

>It's very easy to say "we should just have smarter algorithms", especially if we expect that someone other than ourselves will be writing those algorithms, but I have yet to see a proposal that's sanely implementable, high performance, and low load (even considering the fact that you'd have to completely overhaul the socket API and app the apps using it before you could deploy it).

I think Multipath TCP fills that gap, as it is transparent to applications, unless they start directly dealing with IPv4/IPv6 addresses. I think the combination of Multipath TCP and happy-eyeballs at initial MPTCP connection establishment (to bring up multiple subflows concurrently) would be even better.