Re: [netext] I-D Action: draft-ietf-netext-logical-interface-support-03.txt

[Sri Gundavelli <[e-mail]>]

Hi Raj:




On 9/8/11 9:01 AM, "[e-mail]" <[e-mail]>
wrote:

> 
> Sri,
> 
> Thanks for the reference.
> A few questions from the configuration of the MN in the experiment
> mentioned in the paper:
> 
> 1. Are multiple interfaces active at the same time (eth0 and eth1)?
> 

Yes. Both the scenarios, keeping or both the interfaces active, was tested
by Sawako. 


> 2. How does the bridge interface handle the scenario when you have
> multiple interfaces simultaneously active and operational? From a PMIP6
> standpoint, does the MN (assuming attachment to different MAGs) obtain
> different prefixes from the MAGs?
> 

The brctl mode puts the interfaces in a bridge mode. For all practical
purposes the virtual interface that is configured aboves see's both the
prefixes as prefixes hosted on a link associated with that virtual
interface.



> 3. How does SLAAC work when the bridge IF is used? Given that you have
> different EUI48s associated with Eth0 and Eth1, what does the br IF use
> for address config?
> 

This is a good question. I remember some discussion on this, now I don't
remember. I have to ask Sawako, what is the identifier that she noticed in
her lab, if its a generated id, or if was of one of the physical interface.
We will check on that.

But, it was tested by a Japanese student under Ryuji, and not by me. So no
short cuts, no vaporware, its real and you should be rest assured, it really
worked like spot less :)



Regards
Sri