I've also read both documents.

Yoav Nir <[email protected]> wrote:
    > itself). To my mind a VPN is not an NBMA, but that does not mean an
    > NBMA cannot serve as a good model for a VPN.

I agree.  I like the idea of thinking of it as an NBMA, but not the idea of
using the layer-2 protocol *directly*.  It is clearly a quick (but
Brilliant!) hack inside IOS to reuse some layer-2 ATM code over a WAN.

I also do not like the lack of policy control that GRE/IPsec implies,
specifically it directly leads to LPM only routing, and your comment:

    > natural progression from hub&spoke to mesh. There doesn't seem to be a
    > place for policy on whether a shortcut should or should not be
    > established.

This is really a direct result.
It's not clear to me what prevents parts of the mesh from impersonating other
parts,  etc.

======

I also read: draft-mao-ipsecme-ad-vpn-protocol and while conceptually I found
it okay, I think that the protocol should be inside IKE.

--
Michael Richardson <[email protected]>, Sandelman Software Works


Attachment: pgprjsXLRvFmf.pgp
Description: PGP signature

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

Reply via email to