[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [MEXT] Call for adoption of draft-premec-mext-extended-home-link



You are not wrong. But the two PDP contexts will be related to two different DSMIPv6 sessions, i.e. you have at the same time a IPv4-only home link and IPv6-only home link

 

Gerardo

 

From: Behcet Sarikaya [mailto:behcetsarikaya at yahoo.com]
Sent: Thursday, March 12, 2009 12:20 PM
To: Giaretta, Gerardo; Julien Bournelle; George Tsirtsis
Cc: mext
Subject: Re: [MEXT] Call for adoption of draft-premec-mext-extended-home-link

 

Hi Gerardo,

  I found this in 23.060:

For every IMSI, zero, one, or more dynamic PDP addresses per PDP type can be assigned. For every IMSI, zero, one, or more static PDP addresses per PDP type can be subscribed to

 

this to me says that more than one PDP context, e.g. one for IPv4 another for IPv6 can be established.

 

Am I wrong?

 

Behcet

 

 


From: "Giaretta, Gerardo" <gerardog at qualcomm.com>
To: Julien Bournelle <julien.bournelle at gmail.com>; George Tsirtsis <tsirtsis at googlemail.com>
Cc: mext <mext at ietf.org>
Sent: Thursday, March 12, 2009 12:45:41 PM
Subject: Re: [MEXT] Call for adoption of draft-premec-mext-extended-home-link

Hi Julien,

Without going into 3GPP details, due to backward compatibility issue the default bearer in EPS/SAE can be IPv4-only or IPv6-only (they can also be IPv4v6 in case the legacy networks are upgraded). Therefore the home link may be either IPv4-only, IPv6-only or IPv4v6. Let me know if you are interested in more details.

Cheers
Gerardo

> -----Original Message-----
> From: mext-bounces at ietf.org [mailto:mext-bounces at ietf.org] On Behalf Of Julien
> Bournelle
> Sent: Thursday, March 12, 2009 9:20 AM
> To: George Tsirtsis
> Cc: mext
> Subject: Re: [MEXT] Call for adoption of draft-premec-mext-extended-home-link
>
> Hi george, all,
>
>  thanks for your answer. See below
>
> On Thu, Mar 12, 2009 at 1:25 PM, George Tsirtsis
> <tsirtsis at googlemail.com> wrote:
> > The main use case for this as far as I remember has been 3GPP type
> > links. For reasons that are obscure to me and which others may be able
> > to articulate better, a 3GPP link between say an MN and its HA, can be
> > an IPv6-only link.
> >
> > The 3GPP layer, unfortunately is not entirely agnostic to what goes
> > over it, which is what causes this somewhat  strange case. In other
> > words a 3GPP network can be designed/configured to only support IPv6,
> > in which case there is no way to run native IPv4 over the 3GPP link,
> > even if either side of the link (e.g.  MN and HA) are IPv6 capable.
>
>  ok I see. But what is the interest, for an operator, to configure the 3GPP
> link
> as v6-only if it knows that its clients need v4 ?
>
>  Regards,
>
>  Julien
>
> >
> > I hope this helps
> > George
> >
> > On Thu, Mar 12, 2009 at 11:43 AM, Premec, Domagoj
> > <domagoj.premec at siemens.com> wrote:
> >> Julien,
> >>
> >>>  but what I don't understand is that in your I-D, the MN has
> >>> a v4 HoA, the HA has a v4 address and you are on the home
> >>> link. So I don't see what prevents you from using IPv4 on
> >>> this link. I fail to understand the use case.
> >>
> >> The home link is not a dual stack link and the HA doesn't have the v4
> address on the interface facing the MN and the MN's v4 home address is not
> assigned to the physical interface either. The physical link is the IPv6-only
> and the v4 address can not be used directly, any v4 traffic is tunneled by the
> MN to the HA's IPv6 address.
> >>
> >>>  Do you have a specific deployment in mind ?
> >> Nothing specific really, I would expect this feature to be an interesting
> option for some 3GPP deployments.
> >>
> >> domagoj
> >>
> >> P.S.
> >> The support for the IPv4-only home link was added recently to the DMSIP6
> draft and so the next revision of this draft  focuses only on the IPv6-only
> home link.
> >>
> >>
> >>> -----Original Message-----
> >>> From: Julien Bournelle [mailto:julien.bournelle at gmail.com]
> >>> Sent: 12. ozujak 2009 12:15
> >>> To: Premec, Domagoj
> >>> Cc: marcelo bagnulo braun; mext
> >>> Subject: Re: [MEXT] Call for adoption of
> >>> draft-premec-mext-extended-home-link
> >>>
> >>> Hi,
> >>>
> >>>  see inline,
> >>>
> >>> On Thu, Mar 12, 2009 at 11:51 AM, Premec, Domagoj
> >>> <domagoj.premec at siemens.com> wrote:
> >>> > Hi Julien,
> >>> >
> >>> > The main motivation behind the IPv6-only home link is not
> >>> the inability of the link to carry the IPv4.  The operator
> >>> may choose to deploy the IPv6-only network to save the
> >>> additional managment costs (and headache) associated with
> >>> operating the dual stack network. Or the operator may prefer
> >>> IPv6 due to the IPv4 address shortage. Please have a look
> >>> also at the
> >>> http://www.ietf.org/internet-drafts/draft-ietf-softwire-dual-s
> >>> tack-lite-00.txt for another example of the IPv6-only link.
> >>> Given the above considerations and the need to move towards
> >>> the IPv6 in general, I believe that there is some value in
> >>> the concept of the IPv6-only home link and that there will be
> >>> deployments making use of this capability.
> >>> >
> >>>
> >>>
> >>>  but what I don't understand is that in your I-D, the MN has
> >>> a v4 HoA, the HA has a v4 address and you are on the home
> >>> link. So I don't see what prevents you from using IPv4 on
> >>> this link. I fail to understand the use case.
> >>>
> >>>  Do you have a specific deployment in mind ?
> >>>
> >>>  regards,
> >>>
> >>>  Julien
> >>>
> >>> > regards,
> >>> > domagoj
> >>> >
> >>> >
> >>> >> -----Original Message-----
> >>> >> From: mext-bounces at ietf.org [mailto:mext-bounces at ietf.org]
> >>> On Behalf
> >>> >> Of Julien Bournelle
> >>> >> Sent: 11. ozujak 2009 18:44
> >>> >> To: marcelo bagnulo braun
> >>> >> Cc: mext
> >>> >> Subject: Re: [MEXT] Call for adoption of
> >>> >> draft-premec-mext-extended-home-link
> >>> >>
> >>> >> Hi all,
> >>> >>
> >>> >>  I have a question regarding this draft.
> >>> >>
> >>> >>  What sort of link can carry IPv6 but not IPv4 ?
> >>> >>
> >>> >>  Regards,
> >>> >>
> >>> >>  Julien
> >>> >>
> >>> >> On Wed, Mar 4, 2009 at 9:38 PM, marcelo bagnulo braun
> >>> >> <marcelo at it.uc3m.es> wrote:
> >>> >> > Hi,
> >>> >> >
> >>> >> > This is call for adoption for
> >>> >> draft-premec-mext-extended-home-link-00
> >>> >> > as WG item to cover the charter item:
> >>> >> >
> >>> >> > (A.9) Extended DSMIPv6 Home Network Support: DSMIPv6
> >>> >> assumes the home
> >>> >> > network to be dual stack providing simultaneous IPv6 and
> >>> >> > IPv4 network access. It is proposed to extend DSMIPv6 to
> >>> >> support home
> >>> >> > networks which provides IPv4, or IPv6 respectively,
> >>> direct network
> >>> >> > access only, but where virtual IPv6 home network
> >>> connectivity, or
> >>> >> > virtual IPv4 home network connectivity respectively, may be
> >>> >> obtained
> >>> >> > by tunneling to the HA. The latter shall be obtained by DSMIPv6
> >>> >> > operation using the v4HoA address as Care-of-address for
> >>> the v6HoA
> >>> >> > address, and vice versa, the v6HoA address as
> >>> >> care-of-address for the
> >>> >> > v4HoA address.
> >>> >> >
> >>> >> > The document can be found at:
> >>> >> >
> >>> >> >
> >>> http://tools.ietf.org/html/draft-premec-mext-extended-home-link-00
> >>> >> >
> >>> >> > Please express your opinion on whether this should become a WG
> >>> >> > document before the march 19
> >>> >> >
> >>> >> > thanks, Julien and marcelo
> >>> >> > _______________________________________________
> >>> >> > MEXT mailing list
> >>> >> > MEXT at ietf.org
> >>> >> > https://www.ietf.org/mailman/listinfo/mext
> >>> >> >
> >>> >> _______________________________________________
> >>> >> MEXT mailing list
> >>> >> MEXT at ietf.org
> >>> >> https://www.ietf.org/mailman/listinfo/mext
> >>> >>
> >>> >>
> >>> >
> >>>
> >> _______________________________________________
> >> MEXT mailing list
> >> MEXT at ietf.org
> >> https://www.ietf.org/mailman/listinfo/mext
> >>
> >
> _______________________________________________
> MEXT mailing list
> MEXT at ietf.org
> https://www.ietf.org/mailman/listinfo/mext
_______________________________________________
MEXT mailing list
MEXT at ietf.org
https://www.ietf.org/mailman/listinfo/mext