Re: Re: FW: I-D ACTION:draft-dupont-ikev2-haassign-01.txt

Francis Dupont wrote:
>  In your previous mail you wrote:
>
>    Could you explain some major differences between your solution and
>    this one: the Mobile Node use DHAAD to obtain HA address, then use
>    IKEv2 with cookie step to bootstrap from the HA.
>    
> => simple (and it is the difference between discovery and assignment):
> in our solution the HA side has the control about which HA will be
> used.

In the pure non-anycast rfc3775 DHAAD solution too, the HA side has the 
control about which HA will be used.  The HAs advertise their 
preferences in RA and the MN receives an ordered list of HAs to use. 
The MN will use the HA with the highest preference.  The HAs set their 
preference as they wish, so collectively they know which HA will be used.

Alex