[v6ops] Re: Gorry Fairhurst's Discuss on draft-ietf-v6ops-framework-md-ipv6only-underlay-27: (with DISCUSS and COMMENT)

Gorry Fairhurst <gorry@erg.abdn.ac.uk> Sat, 22 August 2026 09:11 UTC

Return-Path: <gorry@erg.abdn.ac.uk>
X-Original-To: v6ops@mail2.ietf.org
Delivered-To: v6ops@mail2.ietf.org
Received: from localhost (localhost [127.0.0.1]) by mail2.ietf.org (Postfix) with ESMTP id 6ACF812DBA2A4; Sat, 22 Aug 2026 02:11:32 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=ietf.org; s=ietf1; t=1787389892; bh=Pde9wOWCkcvFanlaSOV3tFul5uJf1vBStz++colTDI0=; h=Date:Subject:To:Cc:References:From:In-Reply-To; b=bHWtHvxZBLzd7Bz7CIiaG/o136qA3qmSow4sAlyM95YvFlOrOLq1PNxptiIXCuFP+ ISfbTzPmHD/tf6fxTyJY0fzVcFIr3G76PVvc+EM0iS6dUItPET0LVq5agAmmXdvc1Q aut2FlGIIsU3I8egEJc6kmb+tJkqEE0a03zq/NeU=
X-Virus-Scanned: amavisd-new at ietf.org
X-Spam-Flag: NO
X-Spam-Score: -1.997
X-Spam-Level:
X-Spam-Status: No, score=-1.997 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_VALIDITY_RPBL_BLOCKED=0.001, RCVD_IN_VALIDITY_SAFE_BLOCKED=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Authentication-Results: mail2.ietf.org (amavisd-new); dkim=pass (2048-bit key) header.d=erg.abdn.ac.uk
Received: from mail2.ietf.org ([166.84.6.31]) by localhost (mail2.ietf.org [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ayfakUBgMrAb; Sat, 22 Aug 2026 02:11:31 -0700 (PDT)
Received: from pegasus.erg.abdn.ac.uk (pegasus.erg.abdn.ac.uk [137.50.19.135]) by mail2.ietf.org (Postfix) with ESMTP id 5006012DBA29F; Sat, 22 Aug 2026 02:11:29 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=erg.abdn.ac.uk; s=default; t=1787389883; bh=Pde9wOWCkcvFanlaSOV3tFul5uJf1vBStz++colTDI0=; h=Date:Subject:To:Cc:References:From:In-Reply-To:From; b=D0uQ6ud7h4y0rVViOgQqhfHJUJnpXxvZHDjfSD9rP86VukCXriYBiW6lQpbq/oyeL zFl2x6HZQ3RrnM3erTQPSqwu/KkX44Xu9grtszPSROdlRXt5dJZ9j9z5TtCrDb8Ohc ySuJB1yrz+7WINZtu1k+MQ52YIbiHjDDVNoPTLNppsse+8hatn9e5QwdyY0WllPbTd FlIQ3qSr8Z9h1Jlk4gmjfTFjk2kEV69m9BiXK2AAEWaHZJcIwlww3XeIw1rBz12ORq 5ZahO+LTF6BnF/q1rePEet35goRab2vI8B2GX6hcxCEOH1Ka5T5aFaaiOc7oQIGMQW Rga5XRyCAWGzA==
Received: from [192.168.1.130] (fgrpf.plus.com [212.159.18.54]) by pegasus.erg.abdn.ac.uk (Postfix) with ESMTPSA id 8375D1B00ADD; Sat, 22 Aug 2026 10:11:16 +0100 (BST)
Message-ID: <8739bd54-4b8c-4459-b5ec-ab6403a72f65@erg.abdn.ac.uk>
Date: Sat, 22 Aug 2026 10:11:15 +0100
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Content-Language: en-GB
To: Brian E Carpenter <brian.e.carpenter@gmail.com>, The IESG <iesg@ietf.org>
References: <178713902923.624655.16413669049276131039@dt-datatracker-7c6ddbc678-lb5nk> <8a1b6b80-acc1-47b7-93b9-b2390cdf6f00@gmail.com> <6517cded-3acb-41f9-bebb-8590c2cfb943@erg.abdn.ac.uk> <347a8a05-e970-4365-81b7-98f620a77cb0@gmail.com>
From: Gorry Fairhurst <gorry@erg.abdn.ac.uk>
Organization: UNIVERSITY OF ABERDEEN
In-Reply-To: <347a8a05-e970-4365-81b7-98f620a77cb0@gmail.com>
Content-Type: text/plain; charset="UTF-8"; format="flowed"
Content-Transfer-Encoding: 8bit
Message-ID-Hash: HIT7NBNYUFPTR7ZWO5P6G7E5UPEUEJL2
X-Message-ID-Hash: HIT7NBNYUFPTR7ZWO5P6G7E5UPEUEJL2
X-MailFrom: gorry@erg.abdn.ac.uk
X-Mailman-Rule-Misses: dmarc-mitigation; no-senders; approved; emergency; loop; banned-address; member-moderation; header-match-v6ops.ietf.org-0; nonmember-moderation; administrivia; implicit-dest; max-recipients; max-size; news-moderation; no-subject; digests; suspicious-header
CC: draft-ietf-v6ops-framework-md-ipv6only-underlay@ietf.org, ron@bonica.org, v6ops-chairs@ietf.org, v6ops@ietf.org
X-Mailman-Version: 3.3.9rc6
Precedence: list
Subject: [v6ops] Re: Gorry Fairhurst's Discuss on draft-ietf-v6ops-framework-md-ipv6only-underlay-27: (with DISCUSS and COMMENT)
List-Id: v6ops discussion list <v6ops.ietf.org>
Archived-At: <https://mailarchive.ietf.org/arch/msg/v6ops/bEfFcszh0DUQALQLA8tPtg186Sw>
List-Archive: <https://mailarchive.ietf.org/arch/browse/v6ops>
List-Help: <mailto:v6ops-request@ietf.org?subject=help>
List-Owner: <mailto:v6ops-owner@ietf.org>
List-Post: <mailto:v6ops@ietf.org>
List-Subscribe: <mailto:v6ops-join@ietf.org>
List-Unsubscribe: <mailto:v6ops-leave@ietf.org>

On 20/08/2026 23:09, Brian E Carpenter wrote:
> On 21-Aug-26 05:38, Gorry Fairhurst wrote:
>> On 20/08/2026 02:05, Brian E Carpenter wrote:
>>> Hi Gorry,
>> Hello Brian, and thanks for discussing this, where the terminology is 
>> complex,
>>>
>>> <snip>
>>>
>>>> ## DISCUSS (blocking)
>>>>
>>>> As noted in
>>>> https://datatracker.ietf.org/doc/statement-iesg-handling-ballot-positions-20220121/, 
>>>>
>>>> a DISCUSS ballot is a request to have a discussion on the points 
>>>> below; I
>>>> really think that the document would be improved with a change 
>>>> here, but can be
>>>> convinced otherwise.
>>>>
>>>> ### Section 7:  Operational Considerations
>>>> The current text reads:
>>>>    "In addition, section 4.1 and section 5.1 of
>>>>     [RFC7915] give translators a SHOULD-level option to zero the TOS/
>>>>     Traffic Class octet entirely on translation ("administrative
>>>>     bleaching"), which silently kills ECN if an operator enables it 
>>>> for
>>>>     DSCP hygiene."
>>>>
>>>> (1) Is it possible to avoid terms such as "DSCP hygiene", avoid 
>>>> "kills"
>>>> (blocks?) and avoid/clarify that the term "TOS/Traffic Class octet" is
>>>> Historic. 
>>>
>>> No, "Traffic Class" is not obsolete - it's the official name
>>> of the field in RFC 8200 (STD 86). Similarly, TOS is defined
>>> in RFC 791. According to RFC 2474:
>>>
>>>    A replacement header field, called the DS field, is defined, 
>>> which is
>>>    intended to supersede the existing definitions of the IPv4 TOS octet
>>>    [RFC791] and the IPv6 Traffic Class octet [IPv6].
>>
>> Yes, I do agree in the use of "Traffic Class" as per RFC 8200. I 
>> still think that "TOS/Traffic Class octet" is anappropriate and 
>> historic: treating this as a single byte is no longer appropriate, 
>> and I would suggest that remarking a DSCP ought not to necessarilly 
>> update the 2 bit ECN field. Ignoring this in the past, has resulted 
>> in many unwanted pathologies and continuing to be unclear in 2026 
>> seems wrong to me.
>>
>>> so "DS Field" is the version-independent term.
>> I agree, this is one part of the Traffic Class field.
>>>
>>>> I note that RFC 7915 refers to the former ToS Field" and introduces
>>>> ToS semantics, this detail should be clarified in the present 
>>>> document, because
>>>> such a policy is no longer applicable in 2026.
>>>
>>> Gorry, I can't understand that sentence nor the "SHOULD-level" phrase
>>> in the draft.
>>> But a possible rewrite that I think makes sense is:
>>>
>>>     "In addition, sections 4.1 and 5.1 of [RFC7915] recommend that 
>>> translators
>>>      provide a configuration option to overwrite the Differentiated 
>>> Services
>>>      field octet [RFC2474] on translation. If applied, this would 
>>> silently
>>>      nullify ECN."
>>
>> A treatment that remarks the entire byte when remarking a DSCP is not 
>> appropriate UNLESS this has been configured for the former TOS 
>> semantics, which is no longer recommended.
>
> I agree with that of course, but I'm not sure that this document is 
> the right place to address it. Maybe it's time for RFC7915bis? Coud we 
> simply delete the whole "In addition..." paragraph, which would remove 
> the issue from the present draft?

Removing the text (or at the very least caerfully checking it cannot be 
read as contradicting) sounds like a good plan, I'd be happy with that.

I expect RFC 7915 could be usefully updated on this, but as you note 
that is a different activity.

>
>> * The characterisation of 
>> draft-ietf-v6ops-framework-md-ipv6only-underlay as a NAT64 translator 
>> seems like RFC7915 might apply, but then zeroing the ECN field seems 
>> wrong (as per RFC2474 
>> <https://datatracker.ietf.org/doc/html/rfc2474> and RFC3260 
>> <https://datatracker.ietf.org/doc/html/rfc3260> which assert a 6-bit 
>> field, and more recently by RFC9599 and RFC9601.)
>>
>> I suggest the text ought to be more like this:
>>
>> Sections 4.1 and 5.1 of [RFC7915] recommends that "An implementation 
>> of a translator SHOULD support an administratively configurable 
>> option to ignore the former Type of Service (ToS) octet (IPv4) and 
>> always set the IPv6 traffic class (TC) to zero." However,the default 
>> treatment is to copy the 8-bit Traffic Class (comprising the 6-bit 
>> DSCP Field abd the 2-bit ECN Field) to the new packet. When the  
>> translator is at an administrative boundary, the filtering and update 
>> considerations of [RFC2475 
>> <https://datatracker.ietf.org/doc/html/rfc2475>] are applicable. 
>> Inthis case, the translator MUST NOT  treat the IPv4 former TOS Field
>> Traffic Class octet as a single 8-bit field. This is because the 
>> resulting linkage of ECN and Diffserv field propagation between inner 
>> and outer headers is not consistent with the definition of the 6-bit 
>> Diffserv field in[RFC2474 
>> <https://datatracker.ietf.org/doc/html/rfc2474>] and[RFC3260 
>> <https://datatracker.ietf.org/doc/html/rfc3260>] addition, if the 
>> translator is at an administrative boundary, the  filtering and 
>> update considerations of [RFC2475 
>> <https://datatracker.ietf.org/doc/html/rfc2475>] may be
>> applicable."
>
> (Sorry, the collision between HTML content and plain text has mangled 
> all that, but...)
>
> Your suggested text is better than mine, but I now think that removing 
> the whole paragraph is better.
Thanks.
>
>>>>
>>>> (2) I would like to discuss if this I-D can clearly state that 
>>>> other fields in
>>>> the IPv4/IPv6 header ought to be mapped to corresponding values, in 
>>>> particular
>>>> the DiffServ Field (subject to the filtering and update 
>>>> considerations of
>>>> [RFC2475]) and the ECN Field. As per
>>>> https://datatracker.ietf.org/doc/draft-ietf-v6ops-rfc6146-bis/
>>>
>>> Since the draft cites RFC7915 normatively, and RFC7915 cites RFC6146 
>>> normatively,
>>> I'm not sure there is anything more to do. 
>>
>> I did not see any statement on the former ToS Byte or DSCP usage in 
>> RFC6146 and the default  If I missed something let me know.
>
> You have to follow the breadcrumbs to RFC6145 and then RFC7915, the 
> current standard for stateless translation. There we find (for 
> translation to IPv6):
>
>       An implementation of a translator SHOULD
>       support an administratively configurable option to ignore the IPv4
>       TOS and always set the IPv6 traffic class (TC) to zero.  In
>       addition, if the translator is at an administrative boundary, the
>       filtering and update considerations of [RFC2475] may be
>       applicable.
>
> and (for translation to IPv4):
>
>       However, in some IPv4
>       environments, these bits might be used with the old semantics of
>       "Type Of Service and Precedence".  An implementation of a
>       translator SHOULD provide the ability to ignore the IPv6 traffic
>       class and always set the IPv4 TOS Octet to a specified value.  In
>       addition, if the translator is at an administrative boundary, the
>       filtering and update considerations of [RFC2475] may be
>       applicable.
>
> So I think it's already covered and again, the present draft should 
> probably say nothing, because there's nothing new to say.
>
>     Brian

I would be OK with not digging deeper in present draft,  and see that 
Med has added a reminder to the history of RFC7915bis to see if this can 
be considered (so we do not loose this topic).

Hopefully the present draft can now be updated.

Many thanks,

Gorry

>
>>> RFC6146bis will kick in automatically when published.
>>
>> * For an encapsulation mode, please note the ECN treatement in 
>> RFC2473 was updated by RFC6040, as clarified in RFC9601.
>>
>> If this is some other form of underlay (I am unsure), then this ought 
>> to refer to other parts of: RFC9599.
>>
>>>
>>>     Brian
>>>
>> Gorry
>>>>
>>>>
>>>> ----------------------------------------------------------------------
>>>> COMMENT:
>>>> ----------------------------------------------------------------------
>>>>
>>>> ## COMMENTS (non-blocking)
>>>>
>>>> ### Section 7:  Operational Considerations
>>>>
>>>> “Therefore, network providers must handle MTU and fragmentation 
>>>> issues, for
>>>> example, by configuring appropriate MSS clamping," - This is a 
>>>> TCP-specific
>>>> mechanism, it would be super helpful to indicate that this is for 
>>>> "TCP".
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> v6ops mailing list -- v6ops@ietf.org
>>>> To unsubscribe send an email to v6ops-leave@ietf.org
>>
>>