Re: Objection to draft-ietf-6man-rfc4291bis-07.txt

Le 02/03/2017 à 23:48, Mark Andrews a écrit :
>
> In message <6b89579e-62e8-dfb5-070e-b08989b9492a@gmail.com>,
> Alexandre Petrescu writes:
>> Le 02/03/2017  20:25, JORDI PALET MARTINEZ a crit :
>>> I understood from one of your previous emails that 3GPP doesnt
>>> require DHCPv6-PD.
>>
>> Right - if I understand correctly, it does not require DHCPv6-PD on
>> the UE.  Maybe in the core, but not in the UE.
>>
>>> RFC6653, suggest its use, so probably we got confused ourselves
>>> and 3GPP is actually using it.
>>>
>>> 6rd is a different thing, because is actually IPv4 link and IPv6
>>>  is a tunnel, so you typically use DHCP (v4).
>>
>> I agree.
>>
>>> In IPv6-only world (including cellular), the WAN link will be
>>> IPv6-only instead of dual-stack. This is already happening.
>>>
>>> If they dont use it, I guess is the same reason why big vendors
>>> didnt implemented yet RDDNS
>>
>> I wonder if Windows refusing RDNSS doesn't have somehting to do
>> with its Teredo; and extending this logic to smartphones: the lack
>>  of answers to DHCPv6-PD issued from the UE may have something to
>> do with 464xlat presence.
>>
>> Could it be that 464xlat presence on UE makes it that DHCPv6-PD is
>> impossible on same UE?
>
> No.  They are or should be orthogonal.  One is about providing IPv4
> as as service.  The other is about providing IPv6 address space.

I can agree.

With my operator I asked her once 'why not giving DHCPc6-PD to
smartphone' and she replied 'because now we do 464xlat on the
smartphone, and that is sufficient'.  As such it sounds illogical
statement, although I guess s/he has some reasons to say so.

But let me ask differently: maybe it's because 464xlat requires some
kernel version that DHCPv6-PD doestn support, or vice-versa?  Maybe it's
because once you run 464xlat on smartphone it takes control of the
unique default route and DHCPv6-PD couldnt?  Some reason like that?

Has one tried DHCPv6-PD software simultaneosuly with 464xlat? (even
knowing the cellular operator would not reply to DHCPv6 Slicitations).

>>> Cellular world is using 464XLAT, and they use DHCPv6-PD. Check
>>> RFC6877.
>>
>> I guess you mean cellular stations (like BS, not the User
>> Equipment) uses DHCPv6-PD within the fixed network.
>
> No, the UE should be using DHCPv6-PD.  This is how tethered
> equipement should be getting IPv6 address space.

I agree with yuo.

This is why I dont understand the cellular operators dont do DHCPv6-PD.
  And I guess it is so because they are told it's sufficient to run the
64share INFORMATIONAL RFC.

Are the people advising 64share to tethering-willing telecom operators
conscious it's non-interoperable?

Is the 64share advice coming together with the claim there are enough
/64s out there?

Alex

>> Because cellular operator I am aware of replies to DHCPv6-PD
>> requests issued from an User Equipment.
>>
>> Alex
>>
>>>
>>> Saludos, Jordi
>>>
>>>
>>> -----Mensaje original----- De: ipv6 <ipv6-bounces@ietf.org> en
>>> nombre de Alexandre Petrescu <alexandre.petrescu@gmail.com>
>>> Responder a: <alexandre.petrescu@gmail.com> Fecha: jueves, 2 de
>>> marzo de 2017, 20:15 Para: <ipv6@ietf.org> Asunto: Re: Objection
>>>  to draft-ietf-6man-rfc4291bis-07.txt
>>>
>>>
>>>
>>> Le 02/03/2017  20:06, JORDI PALET MARTINEZ a crit :
>>>> I guess because the lack of DHCPv6-PD support, as you already
>>>> indicated?
>>>
>>> But for deity's sake - DHCPv6-PD is available as open source...
>>> it's common software.  Just download and install.  The IETF Specs
>>> suggest it, and the 3GPP specs require it too.
>>>
>>> There must be something else making these router equipment
>>> manufacturers refraining from downloading and installing
>>> DHCPv6-PD.
>>>
>>>> I think majority of the households and business customers get
>>>> configured the CPE by means for DHCPv6-PD
>>>
>>> AFAIK my CPE at home does not use DHCPv6-PD, but something based
>>>  on IPv4, like 6rd.  I think it's something developped in-house
>>> at that particular operator.
>>>
>>> Alex
>>>
>>>>
>>>> Saludos, Jordi
>>>>
>>>>
>>>> -----Mensaje original----- De: ipv6 <ipv6-bounces@ietf.org> en
>>>> nombre de Alexandre Petrescu <alexandre.petrescu@gmail.com>
>>>> Responder a: <alexandre.petrescu@gmail.com> Fecha: jueves, 2
>>>> de marzo de 2017, 20:02 Para: <ipv6@ietf.org> Asunto: Re:
>>>> Objection to draft-ietf-6man-rfc4291bis-07.txt
>>>>
>>>>
>>>>
>>>> Le 02/03/2017  19:54, JORDI PALET MARTINEZ a crit :
>>>>> Actually, is not correct that most use /56 for residential.
>>>>>
>>>>>> From my last review of the survey, worldwide, 22% use /48,
>>>>>>  35% use /56, but there is a lot of ISPs (33%) doing it
>>>>>> wrong and using /64, which of course, we are explaining
>>>>>> them that is wrong. 10% use other sizes.
>>>>
>>>> So how comes that households can get /56s but smartphones no?
>>>>
>>>> Really there must be something there hidden.
>>>>
>>>> Alex
>>>>
>>>>>
>>>>> Full details at:
>>>>>
>>>>>
>> https://labs.ripe.net/Members/jordipaletm/results-of-the-ipv6-deployment-s
>>
>>
>>
>>
urvey
>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>
>>>>>
>>>>>
>>>>
>>>>> Saludos, Jordi
>>>>>
>>>>>
>>>>> -----Mensaje original----- De: ipv6 <ipv6-bounces@ietf.org>
>>>>> en nombre de Mikael Abrahamsson <swmike@swm.pp.se>
>>>>> Organizacin: People's Front Against WWW Responder a:
>>>>> <swmike@swm.pp.se> Fecha: jueves, 2 de marzo de 2017, 19:33
>>>>> Para: Alexandre Petrescu <alexandre.petrescu@gmail.com> CC:
>>>>> <ipv6@ietf.org> Asunto: Re: Objection to
>>>>> draft-ietf-6man-rfc4291bis-07.txt
>>>>>
>>>>> On Thu, 2 Mar 2017, Alexandre Petrescu wrote:
>>>>>
>>>>>> YEs yes, but how much of that /44 is covering the
>>>>>> end-users and how much is reserved for interconnnections?
>>>>>
>>>>> Nothing. It's /44 to the GGSN/SPGW.
>>>>>
>>>>>> Ah great, but I guess few cellular operators (if any?) are
>>>>>> LIRs. Or maybe that's true and I didnt know.
>>>>>
>>>>> You don't know.
>>>>>
>>>>>> If all this were that simple and clearcut - there are
>>>>>> enough /64s out there - then why operators only assign one
>>>>>>  per one end user?
>>>>>
>>>>> Because DHCPv6-PD hasn't been implemented in mobile networks
>>>>>  yet (that I know of). So that's all they can do per 3GPP
>>>>> standards.
>>>>>
>>>>> Residential rollouts, most use /56 per customer.
>>>>>
>>>>>> From the RIRs, you can without further justification get
>>>>>> /48 per "site",
>>>>> so if you show up to RIR and you're LIR and you say "hello,
>>>>> I have 40 million customers and I want to give each customer
>>>>> a /48" then they'll give you a /22 most likely. You're
>>>>> perfectly within your right as an operator to deploy /48 per
>>>>>  customer per current RIR rules that I am aware of.
>>>>>
>>>>> -- Mikael Abrahamsson    email: swmike@swm.pp.se
>>>>>
>>>>> --------------------------------------------------------------------
>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>
>>>>>
>>>>>
>>>>
>>>>>
>>>>>
>>>>>
IETF IPv6 working group mailing list ipv6@ietf.org Administrative
>>>>> Requests: https://www.ietf.org/mailman/listinfo/ipv6
>>>>> --------------------------------------------------------------------
>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>
>>>>>
>>>>>
>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
********************************************** IPv4 is over Are
>>>>> you ready for the new Internet ? http://www.consulintel.es
>>>>> The IPv6 Company
>>>>>
>>>>> This electronic message contains information which may be
>>>>> privileged or confidential. The information is intended to
>>>>> be for the use of the individual(s) named above. If you are
>>>>> not the intended recipient be aware that any disclosure,
>>>>> copying, distribution or use of the contents of this
>>>>> information, including attached files, is prohibited.
>>>>>
>>>>>
>>>>>
>>>>> --------------------------------------------------------------------
>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>
>>>>>
>>>>>
>>>>
>>>>>
>>>>>
>>>>>
IETF IPv6 working group mailing list ipv6@ietf.org Administrative
>>>>> Requests: https://www.ietf.org/mailman/listinfo/ipv6
>>>>> --------------------------------------------------------------------
>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>>>
>>>>>
>>>>>
--------------------------------------------------------------------
>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>> IETF IPv6 working group mailing list ipv6@ietf.org
>>> Administrative
>>>> Requests: https://www.ietf.org/mailman/listinfo/ipv6
>>>> --------------------------------------------------------------------
>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
********************************************** IPv4 is over Are
>>>> you ready for the new Internet ? http://www.consulintel.es The
>>>> IPv6 Company
>>>>
>>>> This electronic message contains information which may be
>>>> privileged or confidential. The information is intended to be
>>>> for the use of the individual(s) named above. If you are not
>>>> the intended recipient be aware that any disclosure, copying,
>>>> distribution or use of the contents of this information,
>>>> including attached files, is prohibited.
>>>>
>>>>
>>>>
>>>> --------------------------------------------------------------------
>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>
>>>>
>>>>
>>>>
IETF IPv6 working group mailing list ipv6@ietf.org Administrative
>>>> Requests: https://www.ietf.org/mailman/listinfo/ipv6
>>>> --------------------------------------------------------------------
>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>>
>>>>
>>>>
>>>>
--------------------------------------------------------------------
>>> IETF IPv6 working group mailing list ipv6@ietf.org
>>> Administrative Requests:
>>> https://www.ietf.org/mailman/listinfo/ipv6
>>> --------------------------------------------------------------------
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
********************************************** IPv4 is over Are you
>>> ready for the new Internet ? http://www.consulintel.es The IPv6
>>> Company
>>>
>>> This electronic message contains information which may be
>>> privileged or confidential. The information is intended to be for
>>> the use of the individual(s) named above. If you are not the
>>> intended recipient be aware that any disclosure, copying,
>>> distribution or use of the contents of this information,
>>> including attached files, is prohibited.
>>>
>>>
>>>
>>> --------------------------------------------------------------------
>>>
>>>
>>>
>>>
IETF IPv6 working group mailing list ipv6@ietf.org Administrative
>>> Requests: https://www.ietf.org/mailman/listinfo/ipv6
>>> --------------------------------------------------------------------
>>>
>>
>>
>>>
>>>
>>>
--------------------------------------------------------------------
>> IETF IPv6 working group mailing list ipv6@ietf.org Administrative
>> Requests: https://www.ietf.org/mailman/listinfo/ipv6
>> --------------------------------------------------------------------
>