Re: [Mip4] I-D Action:draft-ietf-mip4-nemo-v4-base-07.txt

John, Alex,

My understanding is that the base NEMOv4 specification 
(draft-ietf-mip4-nemo-v4-base-07.txt) requires double encapsulation to 
be used for the Foreign Agent CoA mode, but makes it optional for the 
Collocated CoA mode. In other words, in Collocated CoA mode, the spec 
allows (and actually encourages = SHOULD) the simple (bi-directional) 
encapsulation between MR and HA to be used.

The double encapsulation is required in the FA CoA mode, to ensure 
transparency with respect to legacy (RFC3344) Foreign Agent. If simple 
encapsulation only would be used, the FA would need to be able to 
forward a packet addressed to an LFN address while only the MR's HoA is 
present in the FA's visitor list...and the packet would be dropped. The 
draft "FA extensions to NEMOv4 Base" (draft-ietf-mip4-nemov4-fa-01.txt) 
aims at addressing this specific issue by defining extension to FA to 
make it NEMOv4 aware and thus allow the use of simple encapsulation 
(i.e. of the sole FA<->HA tunnel and nor more the additional inner 
MR_HOA<-->HA tunnel).

Can authors confirm this was the intention of the two different specs 
(i.e. the NEMOv4 base spec allowing simple encapsulation for the 
Collocated CoA mode, while a separate doc specifying it for the FA CoA 
mode)?

I have not checked the last version of the base NEMOv4 spec to determine 
if the current text is explicit enough about these aspects. John what do 
you think?

Thanks
Christophe.

Alexandru Petrescu wrote:
> John.zhao wrote:
>> Hi,all In the section 7, to co-located mode MR, it seems that it can 
>> also send traffics based on the optimization and only tunnel once with 
>> the CoA as its source address. Right?
> 
> Hi John, thanks for pointing this.  The intention is indeed to have the
> simple encapsulation (optimization of double encapsulation) described in
> a separate document and not here.
> 
> I would like to know which paragraph in section 7 read made you think
> optimization is used?  I would like to identify it and keep track.
> 
> I could think of the following in section 7:
>> For optimization, the Home Agent SHOULD only encapsulate the packet 
>> with the tunnel header (source IP address set to Home Agent and 
>> destination IP address set to CoA) for Collocated CoA mode.
> 
> Also know that I'm not sure who/how/when can make modifications to the
> document, since it's advanced a bit past the WG discussion.
> 
> Alex
> 
> ______________________________________________________________________
> This email has been scanned by the MessageLabs Email Security System.
> For more information please visit http://www.messagelabs.com/email 
> ______________________________________________________________________
> 
> 

-- 
Mip4 mailing list: Mip4@ietf.org
    Web interface: https://www1.ietf.org/mailman/listinfo/mip4
     Charter page: http://www.ietf.org/html.charters/mip4-charter.html
Supplemental site: http://www.mip4.org/