Last Modified: 2004-06-24
The main scenario is making it possible for a VPN user to move from one address to another without re-establishing all security associations, or to use multiple interfaces simultaenously, such as where WLAN and GPRS are used simultaneously. It is also intended that the extensions produced by the WG would address specific needs related to other work in IETF, such as modification of SCTP end points without renegotiation of the security associations or the movement of IKEv2-based secure connections to enable Mobile IP signaling to take place.
An explicit non-goal is the construction of a fully fledged mobility protocol. In particular, the WG shall NOT develop mechanisms for the following functions:
o Hiding of mobility from transport layer protocols or applications (beyond what already exists through the use of the tunnel mode). In this respect MOBIKE is different from Mobile IP, HIP, and other mobility protocols.
o IP address changes done by third parties (NATs, firewalls etc). In particular, MOBIKE shall not replace or modify IKEv2 NAT traversal function. MOBIKE handles IP address changes initiated by one of the endpoints of the security associations. NAT traversal handles other address changes. MOBIKE should not be tightly coupled with the NAT traversal function, but it is necessary to specify in which cases (if any) they can be used together, and how they interact.
o Opportunistic authentication or other tools for the reduction of configuration effort. The mechanisms specified in this WG are to be designed for the traditional VPN use case only.
o Any optimization of packet routing paths due to mobility.
o Load balancing. Multihoming is supported only in the sense of failing over to another interface; sending traffic over multiple addresses using the same SA is not supported.
o Use of IKEv1.
The goals of the MOBIKE working group are to address the following issues:
(1) IKEv2 mobile IP support for IKE SAs. Support for changing and authenticating the IKE SA endpoints IP addresses as requested by the host.
(2) Updating IPsec SA gateway addresses. Support for changing the IP address associated to the tunnel mode IPsec SAs already in place, so that further traffic is sent to the new gateway address.
(3) Multihoming support for IKEv2. Support for multiple IP addresses for IKEv2 SAs, and IPsec SAs created by the IKEv2. This should also include support for the multiple IP address for SCTP transport. This should also work together with the first two items, i.e those addresses should be able to be updated too.
(4) Verification of changed or added IP addresses. Provide way to verify IP address either using static information, information from certificates, or through the use of a return routability mechanism.
(5) Reduction of header overhead involved with mobility-related tunnels. This is a performance requirement in wireless environments.
(6) Specification of PFKEY extensions to support the IPsec SA movements and tunnel overhead reduction.
|May 04||Submit Reduced Tunnel Overhead Mode for IPsec to IESG for Informational RFC|
|May 04||Publish first draft on 'PFKEY Extension for Address Updates' (issue 6 above).|
|Aug 04||Submit PFKEY Extension for Address Updates document to IESG for Informational RFC|
|Oct 04||Publish first draft on 'IKEv2 Address Update', covering issues 1 to 4 from the above list.|
|Feb 05||Submit IKEv2 Address Update document to IESG for Proposed Standard RFC|
|Feb 05||Publish first draft on 'Reduced Tunnel Overhead Mode for IPsec' (issue 5 above).|
IKEv2 Mobility and Multihoming WGIETF 60, 2004-08-02 19:30-
Seabreeze room, Sheraton San Diego Hotel & Marina, San Diego, CA
Chairs: Jari Arkko & Paul Hoffman
Scribes: Wassim Haddad, Hannes Tschofenig, Tero Kivinen, and Gabriel Montenegro
Minutes edited by: Pasi Eronen
1. Agenda bashing by chairs
2. MOPO-IKE by Pasi EronenPasi presented draft-eronen-mobike-mopo-00 (see slides). Discussion was postponed to item 4.
3. IKEv2 address management by Francis DupontFrancis presented draft-dupont-ikev2-addrmgmt-05 (see slides).
4. Design discussion by Tero KivinenBasic scenarios:
- Roaming laptop
- Multihoming SGW
- Can be combined
3rd party bombing
How much protection we offer against 3rd party bombing?
- almost none
- do we care if A is the attacker?
NATs and MOBIKE
- Related to 3rd pary bombing issue
- Can't have full 3rd party bombing protection and still work with NATs.
- Only partial protection possible in this case
- Full protection might be downgraded
- Does limited/partial offer that much compared to almost none?
- Should we upgrade protection offered by IKEv2 NAT-T to partial/limited?
- Implicit address update is not manded in IKEv2, only SHOULD
Problems with multihoming and NAT
Case 1. The host behind NAT is not multihoming and the other end is multihoming
- Option 1: Recovery is limited and done only by the host behind NAT.
- Option 2: The host behind NAT must send keepalives to all possible path combinations, and keep the mappings in NAT active all the time
Case 2: The host behind NAT is multihoming, with some of the interfaces using NAT and some not.
- Same problems with interfaces using NAT
- No problems with interfaces not using NAT, can use normal MOBIKE methods.
NAT-T and MOBIKE options
1. Always use NAT-T
- No multihoming in server
- No protection against 3rd party bombing
2. NAT-T and MOBIKE are separate
- If you move to NAT-T, just create new IKE SA which uses NAT-T
- MOBIKE will have the active attack detecter which notices there is a NAT between.
3. NAT-T and MOBIKE are combined
- Modify NAT-T and/or create combination using NAT-T and MOBIKE.
Pasi: I would phrase this a bit differently: can you enable or disable NAT traversal without creating a new IKE SA? If you can, that's option 3. Option 2 also combines NAT-T and MOBIKE since MOBIKE will include NAT detection payloads.Combined NAT-T and MOBIKE
Do NATs only appear when IP address or link status changes?
Do we want to switch back from NAT-T to MOBIKE?
- Save bandwidth
- Better protection against 3rd party bombing
- Attacker can force to use NAT-T
IKEv2 NAT-T is not enough for us
- Can be enabled only at beginning
- Implicit address update not mandatory
- RR checks not mandatory
- No detection of NAT disappearing
Recovery from problems
- Which problems we try to recover
- Only local problems (IP address changes, link down, network card removed)
- Also problems in the network (link breaking down somewhere along the path, routing infrastructure problems, etc.)
- Do we need to act on information that no packets are received from other end?
- Related to NAT-T as there only initiator can fix things
Scope of SA Changes: do all IPsec SAs move together with a common IKE SA?
- Real simultaneous movement where rendezvous server is needed are outside of the scope.
- If we want to recover from some link failures, we will see "virtual simultaneous movements"
- I.e. both ends IP-addresses change at the same time (but to already known addresses)
IPv4 versus IPv6
- Everything should still work.
- We are not discussing of changing the traffic selectors here.
- SGW has only one address
- SGW's addresses are static/mostly static
- No NATs on SGW side
- Allow client to use NAT-T only when not multihoming; i.e. only allow one interface is NAT-T is used.
- If initiator is behind NAT, it takes care of recovery
- Initiator always takes care of recovery
5. Using MOBIKE over NATs by Mohan ParthasarathyMohan presented draft-mohanp-mobike-nat-00 (see slides).
6. Wrap up by chairs(minutes from this section are grouped by topic, not necessarily chronological)
Jari: There was also some support for "full"Interworking with NATs
- Affected by the bombing issue
- Three options for NAT support:
1. Just use NAT-T
2. NAT-T and MOBIKE are separate, use exactly one
3. Some level of NAT-T/MOBIKE integration
- more discussion needed
- We need to agree on NAT terminology.
Recovering from local vs. path failures
- Some support for being able to recover from both
Scope of SA changes
- Disagreement on whether all or only some are moved
- Real simultanous movements out of scope, but "virtual" simultaneous movements needs to be handled
- Should work
- Some simplfiying assumptions are probably going to be needed
Moving forward on the design issues:
- Discussion & consensus calls on the list