Re: [v6ops] Test for adoption as a working group document - Re: I-D Action: draft-binet-v6ops-cellular-host-requirements-02.txt

[Alexandru Petrescu <[e-mail]>]

Le 01/02/2013 20:26, Vízdal Aleš a écrit :
>> -----Original Message----- From: Alexandru Petrescu
>> [mailto:[e-mail]] Sent: Friday, February 01,
>> 2013 6:35 PM To: Vízdal Aleš Cc: [e-mail] Subject: Re:
>> [v6ops] Test for adoption as a working group document - Re: I-D
>> Action: draft-binet-v6ops-cellular-host-requirements-02.txt
>>
>> Le 01/02/2013 12:07, Vízdal Aleš a écrit :
>>>> -----Original Message----- From: [e-mail]
>>>> [mailto:[e-mail]] On Behalf Of Alexandru
>>>> Petrescu Sent: Thursday, January 31, 2013 2:47 PM To:
>>>> [e-mail] Subject: Re: [v6ops] Test for adoption as a
>>>> working group document - Re: I-D Action:
>>>> draft-binet-v6ops-cellular-host-requirements-02.txt
>>>>
>>>> HEllo,
>>>>
>>>> This draft had a competitor in the same space IIRC?  What was
>>>> the name of that draft?
>>>
>>> You're talking about draft-ietf-v6ops-rfc3316bis-00. These drafts
>>> are not competing as each of them has a different goal.
>>
>> I don't understand?
>>
>> The abstract of one says: "This document lists a set of
>> IPv6-related requirements to be supported by cellular hosts."
>>
>> and the abstract of the other includes: "This document considers
>> IPv6 for cellular hosts that attach to the General Packet Radio
>> Service (GPRS), Universal Mobile Telecommunications System (UMTS),
>> or Evolved Packet System (EPS) networks (Hereafter collectively
>> referred to as 3GPP networks)."
>
> Section 1.1 provides the following explanation
>
> This document lists the required features while
> [I-D.ietf-v6ops-rfc3316bis] is doing a good job in identifying
> issues and explaining how to implement basic IPv6 features in a
> mobile context.  Some of the features discussed in
> [I-D.ietf-v6ops-rfc3316bis] are also listed in this document as a
> requirement: the main reason is to collect in one single document a
> comprehensive list of requirements with the required language.

Thanks for pointing to that text.

I need to better understand it.  But I doubt [I-D.ietf-v6ops-rfc3316bis]
is _not_ trying to also have a comprehensive list of requirements.

I have some comments about this comprehensive list of requirements and I
would like to post them here in the v6ops list.  But not sure on which
draft will they go.

>> As an example which makes me think they compete, in the running
>> text one draft says 'should' Prefix Exclude Option (of PD) and the
>> other says MUST for the same.  Both these recommendations are for
>> cellular-enabled Hosts, right?
>>
>>>> In this draft, there is a section about which I have high
>>>> interest - "4. Cellular Devices with LAN Capabilities".  This
>>>> approaches very much to what an IPv6 Mobile Router is, whose
>>>> one particular interface is cellular and others are LAN.
>>>>
>>>> In this respect, implementing an IPv6 Mobile Router there are
>>>> several possibilities and each may have an impact on that
>>>> cellular interface.
>>>>
>>>> For example, the use of Prefix Delegation has an impact on the
>>>> cellular interface - and that is already said as REQ#27.
>>>
>>>> But there are others.  For example, the use of IPv6 Network
>>>> Prefix Translation, or, not least, the use of Mobile IPv6 (with
>>>> NEMO extensions if possible).  This latter is all the more
>>>> important since it is mentioned in the IPv6 Node Requirements
>>>> document, and this draft also mentions that in addition to the
>>>> cellular interface there is a WiFi interface - only Mobile IPv6
>>>> is able to make handovers between the two interfaces.
>>>
>>> It's a scope question. Can you please review the draft and
>>> provide us with your comments?
>>
>> Yes, in section "4. Cellular Devices with LAN Capabilities" lists
>> a number of requirements for Cellular Devices with LAN
>> Capabilities.  Some of these requirements request that that
>> cellular-enabled device uses DHCP Prefix Delegation in order to
>> obtain a Prefix for the LAN.
>>
>> But there exist other methods to achieve the same effect as
>> assigning a Prefix on the LAN from DHCP-PD: Network Prefix
>> Translation, and Prefix Delegation with Network Mobility.
>>
>> My comment is - why aren't these mentioned in the list of
>> requirements?
>
> There has been no demand for anything else than DHCP-PD so far.

Ok, I didn't know that.

Running DHCP-PD on the real interface is one way of obtaining addresses
for the LAN interface.  One other is to run DHCP-PD on a virtual tunnel,
whose other end is the Home Agent of (Mobile IP).  This way is specified
in RFC6276 "DHCPv6 Prefix Delegation for Network Mobility (NEMO)".  This
requires the use of protocol Mobile IP.  The protocol Mobile IP is
necessary when this Router has an additional egress WiFi interface (in
addition to the ingress LAN WiFi interface).

For the same problem (make addresses for devices on the LAN interface),
instead of using DHCP-PD, one may use NPTv6 "Network Prefix Translation"
RFC6296.

Alex

>
>> Regards,
>>
>> Alex
>>
>>>
>>>> Regards,
>>>>
>>>> Alex
>>>
>>> Ales
>
> Ales
>