Re: [v6ops] I-D Action: draft-chen-v6ops-ipv6-roaming-analysis-02.txt

["cb.list6" <[e-mail]>]

Folks,

As a listed author, i am sorry for not having reviewed this document
prior to publication.  I probably should not be on the author list for
this reason :)

Nonetheless, this is a VERY important topic.

First, regarding the draft.  Failure case #5 does not exist since a
home routed PDP terminates on the home network GGSN so the visited
network does not need to deploy NAT64/DNS64.  The roaming user is only
exposed to the home NAT64/DNS64 and all IP packets are GTP tunneled to
the home.

Second, i do not recommend anyone attempt dual-stack (v4v6) 3GPP
roaming.  I am convinced that it is so broken that it may take more
than 5 years to fix.  Essentially, we need to wait of broken gear run
by disinterested networks to simply cycle out.  In fact, the 3GPP
roaming and home serving architectures are so entwined, that i cannot
deploy dual-stack in a 3GPP home network because this will result in
all subscribers to fail for all access attempts in these flawed
roaming partner networks.  This is not acceptable to the business.

It is truly a terrible situation where a roaming partner's flawed
deployment inhibits my ability to do dual-stack in my own network.

That said, there is only one path that i know of to thread this
needle.  If a mobile network operator wants to deploy IPv6 and NOT
break roaming, the mobile network operator MUST deploy an IPv6-only
solution in the home network (such as RFC6877 / 464XLAT) AND require
the phone / UE to request _IPv6-only while at home_ AND _IPv4-only
while roaming_.  The fundamental issue is that widely deployed gear
fails to handle v4v6 permissions correctly, and the mere presence of
this permission causes the user to fail to access anything at all.

I know 464XLAT at 3GPP home and IPv4-only while roaming works, because
i have deployed this.   I know that i can NOT deploy v4v6 because i
deployed that too, and had to roll it back when customers started
failing.

CB

On Mon, Nov 4, 2013 at 9:50 AM, GangChen <[e-mail]> wrote:
> Wg,
>
> We submit the new draft of IPv6 roaming. Please kindly check.
> Your comments are appreciated
>
> BRs
>
> Gang
>
> 2013/11/5, [e-mail] <[e-mail]>:
>>
>> A New Internet-Draft is available from the on-line Internet-Drafts
>> directories.
>>
>>
>>       Title           : IPv6 Roaming Behavior Analysis
>>       Author(s)       : Gang Chen
>>                           Hui Deng
>>                           Dave Michaud
>>                           Jouni Korhonen
>>                           Mohamed Boucadair
>>                           Vizdal Ales
>>                           Cameron Byrne
>>       Filename        : draft-chen-v6ops-ipv6-roaming-analysis-02.txt
>>       Pages           : 12
>>       Date            : 2013-11-04
>>
>> Abstract:
>>    This document intends to enumerate failure cases when a IPv6
>>    subscriber roams into visited network areas.  The investigations on
>>    those failed cases reveal the causes in order to notice improper
>>    configurations, equipment's incomplete functions or inconsistent IPv6
>>    strategy.
>>
>>
>> The IETF datatracker status page for this draft is:
>> https://datatracker.ietf.org/doc/draft-chen-v6ops-ipv6-roaming-analysis
>>
>> There's also a htmlized version available at:
>> http://tools.ietf.org/html/draft-chen-v6ops-ipv6-roaming-analysis-02
>>
>> A diff from the previous version is available at:
>> http://www.ietf.org/rfcdiff?url2=draft-chen-v6ops-ipv6-roaming-analysis-02
>>
>>
>> Please note that it may take a couple of minutes from the time of
>> submission
>> until the htmlized version and diff are available at tools.ietf.org.
>>
>> Internet-Drafts are also available by anonymous FTP at:
>> ftp://ftp.ietf.org/internet-drafts/
>>
>> _______________________________________________
>> I-D-Announce mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/i-d-announce
>> Internet-Draft directories: http://www.ietf.org/shadow.html
>> or ftp://ftp.ietf.org/ietf/1shadow-sites.txt
>>
> _______________________________________________
> v6ops mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/v6ops