Re: IPv6 Deployment for Mobile Subscribers

[Carsten Bormann]

RFC 6177:

   This document obsoletes RFC 3177, updating its recommendations in the
   following ways:

  1) It is no longer recommended that /128s be given out.  While
 there may be some cases where assigning only a single address
 may be justified, a site, by definition, implies multiple
 subnets and multiple devices.

Generally, when you look at an obsolete document such as

https://tools.ietf.org/html/rfc3177

there is a link to the current version ("Obsoleted by: 6177"):

https://tools.ietf.org/html/rfc6177

Do not use websites showing RFCs that do not show this information;
you'll be stuck with outdated specifications.

Grüße, Carsten


Ricardo Ferreira wrote:
> Is there anyone here working in an ISP where IPv6 is deployed?
> We are starting to plan the roll-out IPv6 to mobile subscribers (phones) I
> am interesting in knowing the mask you use for the assignment; whether it
> is /64 or /128.
> 
> In RFC 3177, it says:
> 3. Address Delegation Recommendations
> 
>The IESG and the IAB recommend the allocations for the boundary
>between the public and the private topology to follow those general
>rules:
> 
>   -  /48 in the general case, except for very large subscribers.
>   -  /64 when it is known that one and only one subnet is needed by
>  design.
>   -  /128 when it is absolutely known that one and only one device
>  is connecting.
> 
> Basically a sole device will be connecting to the internet so I am
> wondering if this rule is follwed.
> 
> Cheers
> 

Re: IPv6 Deployment for Mobile Subscribers

[Valdis . Kletnieks]

On Fri, 22 Jul 2016 10:54:48 +0200, Ricardo Ferreira said:
> Is there anyone here working in an ISP where IPv6 is deployed?
> We are starting to plan the roll-out IPv6 to mobile subscribers (phones) I
> am interesting in knowing the mask you use for the assignment; whether it
> is /64 or /128.
>
> In RFC 3177, it says:

>   -  /128 when it is absolutely known that one and only one device
>  is connecting.

See this RFC, which is a recently released BCP:

7934 Host Address Availability Recommendations. L. Colitti, V. Cerf, S.
 Cheshire, D. Schinazi. July 2016. (Format: TXT=37124 bytes) (Also
 BCP0204) (Status: BEST CURRENT PRACTICE) (DOI: 10.17487/RFC7934)

In short - even when you have only one device connecting, you probably need
more than one address.

Also, consider the common practice of tethering


pgpqTrkjoYHfo.pgp
Description: PGP signature

Re: IPv6 Deployment for Mobile Subscribers

[Tore Anderson]

* Baldur Norddahl

> Den 22. jul. 2016 20.25 skrev "Ca By" :
> 
> > Phones, as in 3gpp? If so, each phone alway gets a /64, there is
> > no choice.
> >
> > https://tools.ietf.org/html/rfc6459  
> 
> Here the cell companies are marketing their 4G LTE as an alternative
> to DSL, Coax and fiber for internet access in your home with a 4G
> wifi router. If they can not do prefix delegation it is no
> alternative!

Actually, that /64 prefix is delegated, after a fashion. RFC 7278.

That said, according to RFC 6459 section 5.3, full DHCPv6-PD support
was specified in 3GPP Rel-10. Not sure if there are production
deployments of that yet though, and if not how far off they are. But at
least it looks like it's coming.

Tore

Re: IPv6 Deployment for Mobile Subscribers

[Ryan, Spencer]

> I would love to test it, but it will be no surprise that none of the four
carriers enabled IPv6.


Verizon Wireless has been dual stack for many years, before they ran out of 
public IPv4 addresses and switched handsets to RFC1918 space for v4.


From: NANOG <nanog-boun...@nanog.org> on behalf of Baldur Norddahl 
<baldur.nordd...@gmail.com>
Sent: Friday, July 22, 2016 4:10:41 PM
To: nanog@nanog.org
Subject: Re: IPv6 Deployment for Mobile Subscribers

Den 22. jul. 2016 20.25 skrev "Ca By" <cb.li...@gmail.com>:

> Phones, as in 3gpp? If so, each phone alway gets a /64, there is no
choice.
>
> https://tools.ietf.org/html/rfc6459

Here the cell companies are marketing their 4G LTE as an alternative to
DSL, Coax and fiber for internet access in your home with a 4G wifi router.
If they can not do prefix delegation it is no alternative!

I would love to test it, but it will be no surprise that none of the four
carriers enabled IPv6.

Regards

Baldur

Re: IPv6 Deployment for Mobile Subscribers

[Baldur Norddahl]

Den 22. jul. 2016 20.25 skrev "Ca By" :

> Phones, as in 3gpp? If so, each phone alway gets a /64, there is no
choice.
>
> https://tools.ietf.org/html/rfc6459

Here the cell companies are marketing their 4G LTE as an alternative to
DSL, Coax and fiber for internet access in your home with a 4G wifi router.
If they can not do prefix delegation it is no alternative!

I would love to test it, but it will be no surprise that none of the four
carriers enabled IPv6.

Regards

Baldur

Re: IPv6 Deployment for Mobile Subscribers

[Mikhail Gusarov]

Good day,

On 22 Jul 2016, at 10:54, Ricardo Ferreira wrote:

> Is there anyone here working in an ISP where IPv6 is deployed?

I am not, but I can answer from the consumer's point of view:

> Basically a sole device will be connecting to the internet so I am
> wondering if this rule is follwed.

Tethering.

With best regards,
Mikhail.

Re: IPv6 Deployment for Mobile Subscribers

[Ca By]

On Friday, July 22, 2016, Ricardo Ferreira 
wrote:

> Is there anyone here working in an ISP where IPv6 is deployed?
> We are starting to plan the roll-out IPv6 to mobile subscribers (phones) I
> am interesting in knowing the mask you use for the assignment; whether it
> is /64 or /128.
>
> In RFC 3177, it says:
> 3. Address Delegation Recommendations
>
>The IESG and the IAB recommend the allocations for the boundary
>between the public and the private topology to follow those general
>rules:
>
>   -  /48 in the general case, except for very large subscribers.
>   -  /64 when it is known that one and only one subnet is needed by
>  design.
>   -  /128 when it is absolutely known that one and only one device
>  is connecting.
>
> Basically a sole device will be connecting to the internet so I am
> wondering if this rule is follwed.
>
> Cheers
>
>
Phones, as in 3gpp? If so, each phone alway gets a /64, there is no choice.

https://tools.ietf.org/html/rfc6459


> --
> Ricardo Ferreira
>

Re: IPv6 Deployment for Mobile Subscribers

[Ryan, Spencer]

As far as I'm aware Android still today does not support DHCPv6.


https://en.wikipedia.org/wiki/Comparison_of_IPv6_support_in_operating_systems


From: NANOG <nanog-boun...@nanog.org> on behalf of james machado 
<hvgeekwt...@gmail.com>
Sent: Friday, July 22, 2016 12:57:58 PM
To: Ricardo Ferreira
Cc: NANOG
Subject: Re: IPv6 Deployment for Mobile Subscribers

Ricardo,

I know from previous discussions on this list that Android phones are
looking for DHCPD leases and not /128's or /64's.  From what I remember
this is due to the current requirement for multiple ipv6 subnets for
various applications (vpns among others) to function correctly.  As a
result Google has disabled Android from receiving a DHCP lease as it wasn't
long enough.

if you look back about 6 months there is probably 100+ posts on the subject.

All I really know is that I can not provide an ipv6 dhcp lease to an
android phone and have it receive the address.


james

On Fri, Jul 22, 2016 at 1:54 AM, Ricardo Ferreira <
ricardofbferre...@gmail.com> wrote:

> Is there anyone here working in an ISP where IPv6 is deployed?
> We are starting to plan the roll-out IPv6 to mobile subscribers (phones) I
> am interesting in knowing the mask you use for the assignment; whether it
> is /64 or /128.
>
> In RFC 3177, it says:
> 3. Address Delegation Recommendations
>
>The IESG and the IAB recommend the allocations for the boundary
>between the public and the private topology to follow those general
>rules:
>
>   -  /48 in the general case, except for very large subscribers.
>   -  /64 when it is known that one and only one subnet is needed by
>  design.
>   -  /128 when it is absolutely known that one and only one device
>  is connecting.
>
> Basically a sole device will be connecting to the internet so I am
> wondering if this rule is follwed.
>
> Cheers
>
> --
> Ricardo Ferreira
>

Re: IPv6 Deployment for Mobile Subscribers

[james machado]

Ricardo,

I know from previous discussions on this list that Android phones are
looking for DHCPD leases and not /128's or /64's.  From what I remember
this is due to the current requirement for multiple ipv6 subnets for
various applications (vpns among others) to function correctly.  As a
result Google has disabled Android from receiving a DHCP lease as it wasn't
long enough.

if you look back about 6 months there is probably 100+ posts on the subject.

All I really know is that I can not provide an ipv6 dhcp lease to an
android phone and have it receive the address.


james

On Fri, Jul 22, 2016 at 1:54 AM, Ricardo Ferreira <
ricardofbferre...@gmail.com> wrote:

> Is there anyone here working in an ISP where IPv6 is deployed?
> We are starting to plan the roll-out IPv6 to mobile subscribers (phones) I
> am interesting in knowing the mask you use for the assignment; whether it
> is /64 or /128.
>
> In RFC 3177, it says:
> 3. Address Delegation Recommendations
>
>The IESG and the IAB recommend the allocations for the boundary
>between the public and the private topology to follow those general
>rules:
>
>   -  /48 in the general case, except for very large subscribers.
>   -  /64 when it is known that one and only one subnet is needed by
>  design.
>   -  /128 when it is absolutely known that one and only one device
>  is connecting.
>
> Basically a sole device will be connecting to the internet so I am
> wondering if this rule is follwed.
>
> Cheers
>
> --
> Ricardo Ferreira
>

IPv6 Deployment for Mobile Subscribers

[Ricardo Ferreira]

Is there anyone here working in an ISP where IPv6 is deployed?
We are starting to plan the roll-out IPv6 to mobile subscribers (phones) I
am interesting in knowing the mask you use for the assignment; whether it
is /64 or /128.

In RFC 3177, it says:
3. Address Delegation Recommendations

   The IESG and the IAB recommend the allocations for the boundary
   between the public and the private topology to follow those general
   rules:

  -  /48 in the general case, except for very large subscribers.
  -  /64 when it is known that one and only one subnet is needed by
 design.
  -  /128 when it is absolutely known that one and only one device
 is connecting.

Basically a sole device will be connecting to the internet so I am
wondering if this rule is follwed.

Cheers

-- 
Ricardo Ferreira