ietf
[Top] [All Lists]

RE: [v6ops] Last Call: <draft-ietf-v6ops-mobile-device-profile-13.txt> (An Internet Protocol Version 6 (IPv6) Profile for 3GPP Mobile Devices) to Informational RFC

2014-10-08 10:02:31
Lorenzo,
I have had another look at the list of reqts you highlight. I find each 
referencing a valid RFC or 3GPP section, and each explaining the context in 
which the reqt is valid.
Simply arguing that they do not exist widely today is not reason to discount 
these.
[For your info on privacy extension, the IID is not provided by the network, a 
temp IID is provided during the attach to the PGW but this is then superseded 
by the SLAAC process after the network prefix is learnt via RA. Major handset 
OS including yours provide temporary addresses with pseudo-random IID.]
Regarding DHCPv6 and Prefix Del, then I know of no operator network using this 
today. However this is the only requirement that a mobile operator can give to 
their product marketing people, I feel we would be foolish to lose these. An 
IPv6 requirement that actually could bring new business ☺

May I remind you of the objectives set out in section 1
The objectives of this effort are:

   1.  List in one single document a comprehensive list of IPv6 features
       for a mobile device, including both IPv6-only and dual-stack
       mobile deployment contexts.  These features cover various network
       types such as GPRS (General Packet Radio Service), EPC (Evolved
       Packet Core) or IEEE 802.11 network.

   2.  Help Operators with the detailed device requirement list
       preparation (to be exchanged with device suppliers).  This is
       also a contribution to harmonize Operators' requirements towards
       device vendors.

   3.  Vendors to be aware of a set of features to allow for IPv6
       connectivity and IPv4 service continuity (over an IPv6-only
       transport).

I find the objectives clear and the document goes on to meets those objectives.

In this paper both IPv6-only and dual stack are valid, and the reqts that apply 
to IPv6-only tend to state that; this is an operator choice.
As I understand it:
Verizon Wireless have achieved their fantastic IPv6 penetration using a dual 
stack deployment model (perhaps they would therefore argue that 464xlat is only 
a “should”).
T-Mobile US also have great results with 464xlat on an IPv6-only APN, however 
tethering/wifi hotspot is via another APN.

To provide IPv6-only on a single APN including tethering/Wifi Hostspot is 
difficult due to breadth of terminal support, including some poor tethering 
integration (L_REQ#4 deficiency).
As Orange Poland have embarked on this deployment, a view from Orange Poland 
and terminals would be welcomed!
My perception is that, in this scenario, a number of OS and OEM implementations 
that can support dual stack perfectly well have not achieved suitability here.
For operators who need to decouple their growth from IPv4 addressing, your 
synopsis of “terminals not blocking rollout” is not the reality. Terminals 
remain an issue.
But it is not the intention to list problems but to highlight what is required 
for service continuity.
Regards,
Nick


From: Lorenzo Colitti [mailto:lorenzo(_at_)google(_dot_)com]
Sent: 07 October 2014 13:22
To: BINET David IMT/OLN
Cc: Heatley, Nick; v6ops(_at_)ietf(_dot_)org WG; IETF Discussion; IETF-Announce
Subject: Re: [v6ops] Last Call: <draft-ietf-v6ops-mobile-device-profile-13.txt> 
(An Internet Protocol Version 6 (IPv6) Profile for 3GPP Mobile Devices) to 
Informational RFC

On Tue, Oct 7, 2014 at 8:42 PM, 
<david(_dot_)binet(_at_)orange(_dot_)com<mailto:david(_dot_)binet(_at_)orange(_dot_)com>>
 wrote:
Show me an operator whose rollout is genuinely blocked on terminal features and 
I will believe you. But word from everyone I've talked to is that terminal 
features are not the blocker. Operators such as Verizon Wireless and T-Mobile 
in the US have deployed tens of millions of IPv6-capable devices, and none of 
those devices (and, I'd argue, no commercial devices, anywhere) implement all 
the features in this profile. The vast majority only support a handful.
[DB] Do we consider that all features are mandatory in the draft ? Not at all 
and it demonstrates that RFC 2119 terminology is useful.

Ok, then. So here are examples of musts that are not required for IPv6 
operation in a mobile network, and not supported in widely deployed mobile 
platforms:

C-3 PDP context fallback
C-9 RDNSS
C-10 DHCPv6
C-11 DNS provisioning order
C-12 PDP type limitation
W-1 IPv6-only wifi
A-1 Privacy addresses (because the IID is provided by the network)
A-5 Prefer IPv6 DNS server
L-1 DHCPv6 PD
L-2 Full Customer Edge Router support
A-2 Applications must be IP agnostic
A-3 URI format

Whatever you think, it is still a problem to get some IPv6-ready devices and 
that explains why some on-going IPv6 deployment still rely on some limited 
number of devices in some areas.

Then please explain to me how Verizon Wireless supports IPv6 on all its LTE 
devices, and has done since 2011?

the document does not add any new hurdles for IPv6 deployment since it list 
some requirements based on existing specifications.

But it does add hurdles for IPv6 deployment. Because it lists lots of 
requirements that are not required for IPv6 deployment in mobile networks, and 
that are not widely supported by mobile devices.

We should speak for ourselves and should not imagine how other people will 
consider such document.

Sorry, no. As IETF contributors it is our job to consider what other people 
will think when they read the documents that we produce.

NOTICE AND DISCLAIMER
This e-mail (including any attachments) is intended for the above-named 
person(s).  If you are not the intended recipient, notify the sender 
immediately, delete this email from your system and do not disclose or use for 
any purpose.  
 
We may monitor all incoming and outgoing emails in line with current 
legislation. We have taken steps to ensure that this email and attachments are 
free from any virus, but it remains your responsibility to ensure that viruses 
do not adversely affect you. 

EE Limited
Registered in England and Wales
Company Registered Number: 02382161
Registered Office Address: Trident Place, Mosquito Way, Hatfield, 
Hertfordshire, AL10 9BW
<Prev in Thread] Current Thread [Next in Thread>