Re: [atn] [EXTERNAL] Re: Embedding IP information in an IPv6 address (OMNI) Wed, 14 October 2020 17:45 UTC

Return-Path: <>
Received: from localhost (localhost []) by (Postfix) with ESMTP id AAAE23A0F9C; Wed, 14 Oct 2020 10:45:20 -0700 (PDT)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -1.899
X-Spam-Status: No, score=-1.899 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id acVzmqLDwmJI; Wed, 14 Oct 2020 10:45:19 -0700 (PDT)
Received: from ( []) (using TLSv1.2 with cipher ADH-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by (Postfix) with ESMTPS id 21A693A0F97; Wed, 14 Oct 2020 10:45:18 -0700 (PDT)
Received: from (unknown [IPv6:2a01:79c:cebd:9724:806f:2cf:fdeb:8bfe]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by (Postfix) with ESMTPSA id E37914E11CAC; Wed, 14 Oct 2020 17:45:17 +0000 (UTC)
Received: from [IPv6:::1] (localhost [IPv6:::1]) by (Postfix) with ESMTP id EC96A40A83E3; Wed, 14 Oct 2020 19:45:15 +0200 (CEST)
Content-Type: text/plain; charset=us-ascii
Mime-Version: 1.0 (Mac OS X Mail 13.4 \(3608.\))
Subject: Re: [atn] [EXTERNAL] Re: Embedding IP information in an IPv6 address (OMNI)
In-Reply-To: <>
Date: Wed, 14 Oct 2020 19:45:15 +0200
Cc: =?utf-8?B?56We5piO6YGU5ZOJ?= <>, Robert Moskowitz <>, 6man WG <>, "" <>
Content-Transfer-Encoding: quoted-printable
Message-Id: <>
References: <> <> <> <> <> <> <> <> <> <>
To: "Templin (US), Fred L" <>
X-Mailer: Apple Mail (2.3608.
Archived-At: <>
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: "IPv6 Maintenance Working Group \(6man\)" <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Wed, 14 Oct 2020 17:45:21 -0000

Thanks Fred for the explanation.

>>> What in your opinion would be easier - a) update RFC4291 to allow coding of
>>> the link-local address 54 zero bits, or b) update RFC4861 to allow routers to use
>>> site-local addresses instead of link-local?
>>> We need a good answer for this - either a) or b). The benefit of what is being
>>> proposed by OMNI is too great to simply say no to both.
>> It's unclear to me what the benfits are.
>> Would you be able to summarize here, or point to the relevant paragraphs in the OMNI draft? 
> The benefits include:
> + a means for configuring a unique IPv6 link-local address (i.e., the "OMNI LLA")
>  that is known to be unique on the link without requiring SLAAC or MLD/DAD

Benefit is in the eye of the beholder.
Manual configuration is a common cause of duplicate address. Why can't that happen here?
If you the setup here is that you have "some" information pre-configured. Why not also the link-local address?

> + a means for asserting and/or receiving IPv6 GUA prefixes without explicit
>   prefix delegation messaging over the wire

Implied prefix delegation by the implementation knowing that it should find it's prefix at bit position <n:m> in the packet... doesn't seem like following the principle of least astonishment.
Likewise having the routing system glean information from RSs... is that really the level of adaptiveness you need from a routing system?

I'm not sure I get the motivation between these choices?
As opposed to using existing IPv6 mechanisms for the same.

Best regards,

> So, for example, if a node has been pre-assigned an IPv6 GUA prefix 2001:db8:1:2::/64
> then it can configure the OMNI LLA fe80:2001:db8:1:2:: (but note that this is not RFC4862
> SLAAC; it is a simplified means of setting up an administratively-configured LLA that is
> known to be unique on the link). The node can then assert its IPv6 GUA to the routing
> system by sending an RS with fe80:2001:db8:1:2:: as the source address and with a
> Prefix Length value 'N'.
> In a second example, if a node has no pre-assigned IPv6 GUA then it can send an RS
> message with IPv6 address unspecified. When it gets back an RA, it can determine its
> (delegated) IPv6 GUA prefix by examining the destination address. For example, if
> the RA message has destination address fe80:2001:db8:3:4:: and the message includes
> a Prefix Length 'N' and lifetime value, it can adopt fe80:2001:db8:3:4:: as its own
> OMNI LLA and adopt 2001:db8:1:2::/N as its own IPv6 GUA prefix.
> There are also IPv4-compatible OMNI LLAs of the form fe80::ffff:[v4addr] and
> "Service" OMNI LLAs of the form fe80::[32-bit ID]. All of these address forms
> appear in Section 7 of the document, and their use is discussed throughout:
> Thanks - Fred
>> Best regards,
>> Ole=