Hi Dino,
I have no problems with the control plane part. An encap with multiprotocol support would allow to do IPsec encap before LISP encap, and could be used with the unauthenticated DH mechanism that you propose.

I do really think that the LISP WG should not miss the encap debate, and drive the transition to a format that lends itself to the various use cases that are being envisioned (and that IMO should become the main focus of the WG asap). There's quite a broad support behind VXLAN-GPE, and LISP-GPE is an opportunity for LISP to capitalize on that support and maintain some backward compatibility with the current LISP encap and features.

Fabio

On 12/4/14, 1:30 PM, Dino Farinacci wrote:
Do you support the control-plane mechanisms in the draft?

The advantage of having more bits for encryption is that we can have more 
data-plane keys. With the current lisp-crypto proposal, we only have 3 keys-ids.

But I would use caution here, because the state of the art in data-plane 
encapsulations is all over the place. And while generalization is nice, it 
often isn't practical. And what results in the marketplace is a subset of 
options being delivered in a general design that brings you back to the same 
result of having a specific format that would need to change later.

Dino

On Dec 4, 2014, at 11:08 AM, Fabio Maino <[email protected]> wrote:

I don't support adoption of this document.

The document is proposing an extension of the LISP header to support data plane 
security.  However, there has been quite a lot of discussion in various WGs, 
including LISP, about the need for a more flexible overlay encapsulation. 
Besides support for data plane security, the requirements include capability to 
support non IP payloads, and to support metadata for various applications 
including service chaining  and policy tags.

I believe that rather than just adding incremental support for data plane 
security, the WG should  have a comprehensive look at how to extend the LISP 
header to address the requirements above. draft-lewis-lisp-gpe, that was 
presented to the WG a few times, tries to address all of those requirements.

Extending the semantic of the header to support data plane security, at least 
for HW implementations, comes at about the same cost of addressing all of the 
requirements above. I believe the same is true, to a lesser extent, even with 
SW implementations (at least those that deal with the encap in the kernel).

Thanks,
Fabio



On 12/4/14, 3:27 AM, Luigi Iannone wrote:
Hi All,

During the 91st IETF authors of the draft-farinacci-lisp-crypto-01
[https://tools.ietf.org/html/draft-farinacci-lisp-crypto-01]
asked for WG adoption. Meeting participants expressed consensus on adoption.

This message begins the two weeks call for WG adoption to confirm the meeting 
outcome.
The call ends on  December 19th 2014.

Please respond to the LISP mailing list with any statements of approval or 
disapproval.

Recall that:

- This is not WG Last Call. The document is not final, and the WG is expected to
   modify the document’s content until there is WG consensus that the content 
is solid.
   Therefore, please don’t oppose adoption just because you want to see changes 
to its content.

- If you have objections to adoption of the document, please state your reasons 
why,
   and explain what it would take to address your concerns.

- If you have issues with the content, by all means raise those issues and we 
can
   begin a dialog about how best to address them.
                                                                                
                                                Luigi and Joel

_______________________________________________
lisp mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/lisp
_______________________________________________
lisp mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/lisp

_______________________________________________
lisp mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/lisp

Reply via email to