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

Mikael Abrahamsson <> Tue, 14 October 2014 10:14 UTC

Return-Path: <>
Received: from localhost ( []) by (Postfix) with ESMTP id EF2921A7032 for <>; Tue, 14 Oct 2014 03:14:39 -0700 (PDT)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -2.229
X-Spam-Status: No, score=-2.229 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HELO_EQ_SE=0.35, PLING_QUERY=0.994, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.786, SPF_PASS=-0.001, URIBL_RHS_DOB=1.514] autolearn=ham
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id wslaBL5yovIP for <>; Tue, 14 Oct 2014 03:14:37 -0700 (PDT)
Received: from ( []) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by (Postfix) with ESMTPS id 76E891A702A for <>; Tue, 14 Oct 2014 03:14:37 -0700 (PDT)
Received: by (Postfix, from userid 501) id E3F7EA1; Tue, 14 Oct 2014 12:14:35 +0200 (CEST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple;; s=mail; t=1413281675; bh=z7WfXU8rdmGlkgLExn5uqnqvAWa2PF0tT6JlUWo68bw=; h=Date:From:To:cc:Subject:In-Reply-To:References:From; b=SwCzt6qWCkHZQYe0J72BL3tJWWLTz5BBSI/oBuwWzfYYzufW5utoOdM9ySD1vpcJq FTTjU6N7HZFyJfl5A+TIco1lI0O97YJz15dAhq2qRUegpf3THbC9gAPYNTUhxWTL2/ 7++ch2/ndQo7oFv8BWcmhDF2rDjt2fNbY5do+j9M=
Received: from localhost (localhost []) by (Postfix) with ESMTP id DD4B49F; Tue, 14 Oct 2014 12:14:35 +0200 (CEST)
Date: Tue, 14 Oct 2014 12:14:35 +0200 (CEST)
From: Mikael Abrahamsson <>
To: Pierre Pfister <>
In-Reply-To: <>
Message-ID: <>
References: <> <> <> <> <>
User-Agent: Alpine 2.02 (DEB 1266 2009-07-14)
Organization: People's Front Against WWW
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed
Cc: Erik Kline <>, HOMENET Working Group <>, Markus Stenberg <>
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: Tue, 14 Oct 2014 10:14:40 -0000

On Tue, 14 Oct 2014, Pierre Pfister wrote:

> Looks like a good default policy to me.
> So there always is at least one IPv6 prefix (if not a GUA, generate a ULA).
> It still provides always-on IPv6 connectivity. And would therefore simplify protocol design and implementation.
> Does it seems like a better compromise to you (Mikael, Erik, Wuyts) ?

I would like to see the following (configurable):

If you have GUA, don't use ULA.
When last GUA lease expires, just keep this GUA until something else shows 
up, then deprecate and move away from it.

I would really prefer to have RIO so host doesn't see default route using 
this GUA with no connectivity.

My rationale for this is to avoid to have GUA expire and popping up ULA, 
and then deprecating ULA again when GUA shows up.

Advantage of deprecating GUA as soon as outside connectivity goes away is 
that using 0 lifetime RAs for this prefix when there is no outside 
connectivity and popping up ULA, means RFC6724 hosts will do the right 

I'm conflicted as to what the correct behaviour is. RFC3484 hosts probably 
doesn't support RIO either, so we're screwed either way.

Mikael Abrahamsson    email: