[Mip4] Re: Clarification for Dynamic HA assignment draft

[Alpesh <[e-mail]>]

Pete:

Okay, so I see that you mean. One concern that we had was that Diameter
is still not an RFC and we would be referring to something not a RFC yet.

That said, we don't have a problem in incorporating your observation. We can
still continue using the term ALL_ZERO_ONE_ADDR - with specific meanings
for 0.0.0.0 and 255.255.255.255.

Following is the new text ...

" ALL_ZERO_ONE_ADDR stands for 0.0.0.0 or 255.255.255.255.
255.255.255.255 would indicate assigning  HA in home domain and
0.0.0.0 would mean MN just needs a dynamic
HA, it does not care whether in home or visited domain."

As already mentioned in the framework, the specific mechanisms to assign 
the
local or visited HA are outside the scope.

Let us know what you think.
Alpesh

Pete McCann wrote:

>Hi, Alpesh,
>
>I think we should attempt to be compatible with the functionality
>specified in the Mobile IPv4 Diameter specification, which is inside
>Section 1.4 of draft-ietf-aaa-diameter-mobileip-14.txt:
>
>1.4  Allocation of Home Agent in Foreign Network
>
>   The Diameter Mobile IP application allows a home agent to be
>   allocated in a foreign network, as required in [MIPREQ, CDMA2000].
>   When a foreign agent detects that the mobile node has a home agent
>   address equal to 0.0.0.0 or 255.255.255.255 in the Registration
>   Request message, it MUST add a MIP-Feature-Vector AVP with the Home-
>   Agent- Requested flag set to one. If the home agent address is equal
>   to 255.255.255.255, then the foreign agent also MUST set the Home-
>   Address-Allocatable-Only-in-Home-Realm flag equal to one. If the home
>   agent address is set to 0.0.0.0, the foreign agent MUST set the Home-
>   Address-Allocatable-Only-in-Home-Realm flag equal to zero.
>
>So yes, both values are used to request dynamic home agent, however,
>one of them (255.255.255.255) indicates a preference for home realm
>based HA while the other (0.0.0.0) represents that the MS doesn't care
>whether the HA is in the home or visited realms.
>
>-Pete
>
>
>Alpesh writes:
> > Pete:
> > 
> > This is regarding your comment/observation in last IETF while Kent
> > was presenting the Dynamica HA assignment framework draft.
> > 
> > Currently, we are proposing to use either 0.0.0.0 or 255.255.255.255
> > as the HA address to indicate dynamic HA assignment. I cross-checked
> > IS-835C and they use these addresses interchangeably.
> > 
> > You mentioned that some other draft is trying to separate the meaning of
> > all zero and all ones. My concern is that, it the meaning is not already 
> > defined,
> > should we rely on that draft? We want to accomodate the 3GPP2 model and
> > since they are using both values to mean dynamic HA assignment, we want to
> > go on that path.
> > 
> > In future, if different meanings get assigned to each value, we can 
> > probably change
> > our draft.
> > 
> > Do you have comments/suggestions??
> > 
> > -a
>
>  
>



_______________________________________________
Mip4 mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/mip4