Re: [multimob] draft-ietf-multimob-pmipv6-source-06

[Behcet Sarikaya <[e-mail]>]

Hi Thomas,

My question was not related to this document specifically.
So my conclusion from your reply is that there is no written document on
so-called policy routing in PMIP. Certainly RFC 5213 does not mention it.

Is it fair to say that?

You are referring to some possible email messages, any links?



(I copied Hitoshi who was very interested in this).


BTW we will shepherd your document to IESG very soon.

Regards,


On Mon, Jan 6, 2014 at 4:51 PM, Thomas C. Schmidt <
[e-mail]> wrote:

> Hi Behcet,
>
> I'll shorten the history to make things more readable ...
>
>
> On 06.01.2014 23:22, Behcet Sarikaya wrote:
>
>          Sec. 3.2.4
>>         the multicast
>>              routing information base (MRIB) of the MAG - I thought this
>>         was not
>>         possible? later on seems to be clarified in Sec. 4.3 & 4.4 (I mean
>>         according to our face-to-face conversations in the past?)
>>
>>
>>     In this draft, we describe the setting of an MRIB in agreement with
>>     the topology of the access network. Like a unicast RIB, an MRIB can
>>     of course always express a regular, topology-oriented routing
>>     according to destination addresses in IP headers.
>>
>>     Our discussions, which also took place on the mailing list earlier
>>     and in WG meetings, were concerned with policy-based PMIP routing
>>     according to source addresses. This cannot be expressed in a regular
>>     MRIB.
>>
>>
>> Where did you get this from? Any references?
>>
>>
> Well, the relevant part in the context of this draft is that one can
> express regular, topology-oriented routing in an access network within a
> regular routing table (unicast as well as multicast). It is 'sort of the
> core of IP routing', so I guess there is no doubt about that and a
> reference is not needed (otherwise we would need to recall the fundamentals
> of IP routing).
>
> The other side about policy-based routing in PMIP was discussed some time
> ago, but is not mentioned in the draft. So there is no need to make a
> reference to it.
>
> If you are interested in the discussion, I think major parts have been on
> the list about two years ago. There were contributions from Sri and in
> particular Stig provided longer explanations on what can be possibly done
> and what not.
>
> But this is all outside the current spec (and of any spec that is MULTIMOB
> WG Item), so no need to return to it.
>
> Cheers,
>
> Thomas
>
>
>
> --
>
> Prof. Dr. Thomas C. Schmidt
> ° Hamburg University of Applied Sciences                   Berliner Tor 7 °
> ° Dept. Informatik, Internet Technologies Group    20099 Hamburg, Germany °
> ° http://www.haw-hamburg.de/inet                   Fon: [phone]°
> ° http://www.informatik.haw-hamburg.de/~schmidt    Fax: [phone]°
>