Re: [v6ops] new draft: draft-chen-v6ops-ipv6-roaming-analysis

[Victor Kuarsingh <[e-mail]>]

Authors,

Dave Michaud from our technology team reviewed the document and had some
feedback.  I have included it below.

[Feedback]
Section 2, there should be a mention of HLR
 

Section 3.1 mixes two concepts that have different causes and impacts
-          The UE requesting IPv4v6 PDP/PDN Type

-          The HLR/HSS sending a profile with dual-stack enabled
(extended-PDP-IP parameter)

 
Section 3.2, there is a responsibility that is put on the UE where it
shouldn¹t:
A roaming subscriber with IPv4v6 PDP/PDN type should change the request to
two separated PDP/PDN messages of single IP version in order to achieve
equivalent results.
In reality, the UE will still request IPv4v6 but will be informed by the
network that only single address bearers are allowed (cause code #52)
along with the activation of one address type. Then and only then the UE
can go back and request another bearer with the secondary address type.
 

Section 3.3 a visited network that is IPv6-Only is not something I would
consider probable. Given that the address allocation is the burden of the
home network, I really don¹t see why an operator who wants to eb IPv6-Only
would block inbound roaming using IPv4 or IPv4v6.
 

Section 4.1 makes an assumption about support for IPv6 that has been in
place for as long as IPv4 and it is well supported. Given that this is an
optional feature that must be turned-on on vendor equipment, it¹s not
really the case. But then, this section goes on to describe that a user
could roam into an IPv4-Only network.
 

Section 4.2 I don¹t really understand what one is trying to say but there
is discussion about hard coding a PREFIX64 which could be a mismatch with
the visited network. These functions are all home network based unless
VPLMN local breakout is used but there is no mention of that.
 

Aside from the clarifications above, there should be a clear mention of
local breakout impacts in this document because it shift a large
proportion of the responsibility on the visited network.

 
Also, one thing note mentioned and that we use actively is a clear
separation between what the UE requests and what gets allocated by the
HPLMN. It is not enough to say ³Roaming behavior from a dual-stack
network². Our UEs will be requesting dual-stack to help with compatibility
but our PGW will make a selection (IPv4, IPv6 or IPv4v6). Does this fall
in the category ³from a dual-stack network²?


Regards,

Victor K




On 2013-07-09 8:45 AM, "[e-mail]" <[e-mail]> wrote:

>
>A new draft has been posted, at
>http://tools.ietf.org/html/draft-chen-v6ops-ipv6-roaming-analysis. Please
>take a look at it and comment.
>_______________________________________________
>v6ops mailing list
>[e-mail]
>https://www.ietf.org/mailman/listinfo/v6ops