[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

MIP 4 BOF



MIP 4 BOF

Deployment focus, looking at experience with implementation and deployment of the mobileIP specs and looking at interoperability, progressing base mobile IPv4 specs to draft, completion of AAA key exchange for mobile IP, AAA NAI completion and MIPv4-AAA review, dynamic HA assignment, and VPNs. Milestones enumerated

There appears to be the possibility of giving folk an opportunity to revisit mobile IP and start augmenting the mobile IP model, as distinct from the deployment focus of the charter. There is also ther problem of creep in to other areas. (Tomas Narten did a sterling effort at course correction at the start of the BOF!)

Presentation on mobileip-vpn-problem-statement, a design team output with a security review done by Radia. Wants to last call this and push it to the IESG for publication. The solution update uses an established base, and some as yet uncompleted optimizations. The problem is IPSEC to the DMZ of an enterprise VPN, and the problem of access point mobility, and how do you detect 'inside' and 'outsiode' states. The issue is the IPSEC FA state and the need to regenotiate the state on client movement.

There is a certain amount of complexity creep going on here in order to undertake IPSEC VPN access and mobility where the internal home agent is occluded from external visibility (IPSEC mobility), and its not entirely clear whether these efforts would make mobileIP more attractive to deployment or whether it all becomes yet another reason not to deploy. Charlie Perkins characterized this as "going from the complex to the fantastically complex". Alternatives of using IPSEC tunnel mode as the MoileIP tunnel were discussed, but this scenario was not explicitly included in the current 3 level drafts.

On the whole the concept of using deployment experience to refine the MobileIP specs is a worthy effort, but with the BOF perspective that the extensions with respect to VPNs and IPSEC mobility, the WG would need careful chairing to dampen down the 'charge ahead' enthusiasm and complete the rest of the charter, which in theory is more mindane!.