Re: [homenet] Let's make in-home ULA presence a MUST !?

Ted Lemon <> Mon, 20 October 2014 20:32 UTC

Return-Path: <>
Received: from localhost ( []) by (Postfix) with ESMTP id 9D1B91ACE23 for <>; Mon, 20 Oct 2014 13:32:34 -0700 (PDT)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -0.917
X-Spam-Status: No, score=-0.917 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, PLING_QUERY=0.994, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=no
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id 9HKkv3QUxNjU for <>; Mon, 20 Oct 2014 13:32:33 -0700 (PDT)
Received: from ( []) by (Postfix) with ESMTP id CD7F71ACE25 for <>; Mon, 20 Oct 2014 13:32:33 -0700 (PDT)
Received: from [] ( []) by (Postfix) with ESMTPSA id A1B3E238013F; Mon, 20 Oct 2014 16:32:32 -0400 (EDT)
Content-Type: text/plain; charset=us-ascii
Mime-Version: 1.0 (Mac OS X Mail 7.3 \(1878.6\))
From: Ted Lemon <>
In-Reply-To: <>
Date: Mon, 20 Oct 2014 16:32:27 -0400
Content-Transfer-Encoding: quoted-printable
Message-Id: <>
References: <> <> <> <> <> <> <> <> <> <> <> <> <> <> <> <> <CADhXe53EiyUKM9DBwqVrVQv=ofgw8no4fpz83Dy0rC9UY-4SJQ>
To: James Woodyatt <>
X-Mailer: Apple Mail (2.1878.6)
Cc: HOMENET Working Group <>
Subject: Re: [homenet] Let's make in-home ULA presence a MUST !?
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Mon, 20 Oct 2014 20:32:34 -0000

On Oct 20, 2014, at 4:24 PM, James Woodyatt <> wrote:
> Did you respond to my previous criticism of this idea? If so, then I missed it. It's not a good idea to commission a new standalone network with the same ULA as a previously commissioned network, because it destroys the main property of ULA prefixes that makes them useful: the statistical unlikelihood of merge collisions in the global address realm.

Yes, I read your explanation, and the solution I proposed takes it into account.   Please stop arguing to win and actually read what I wrote.

>> The reason I think it's beneficial is that it reduces to the minimum the number of instances where a long-lived connection will have to be broken because of a renumbering event.   I don't think we can reduce that number to zero, but I think we can make it a lot less likely than it would be if we renumber every time the upstream link goes away.
> Sounds to me like a benefit of very dubious value at best.

Look at it this way: if you have a network and it occasionally gets reconfigured in a way that an application connection _cannot_ survive, that's bad, but we do accept that such situations occur from time to time.   But what if you have a network that is mainly stable, but you have a provider that isn't entirely reliable.   Is this network equivalent to one where you have a monkey who comes in every thirty minutes and swaps all the cables around?

Yes, we have to ride out unusual configuration change events.   But is it really "of very dubious value" to try to minimize the impact these events have on the stable operation of the network?  I don't think it is.

BTW, your flaky radio scenario isn't actually better if you have a GUA; arguably it's worse, particularly since it's probably an LLN that expects flakiness and knows how to ride it out, but does _not_ expect to be renumbered on a regular basis.   Please diagram that scenario and make sure you understand all the state transitions before you use it as a justification for not bothering to try to have a stable ULA.