On Wed, Dec 2, 2009 at 12:02 AM, Brian E Carpenter <[email protected]> wrote: > On 2009-12-02 10:02, Noel Chiappa wrote: > ... >> But if addresses can be added later, then the mechanism for doing so has to >> be secured. (Although there are a number of ways to do that; but if there's >> a MITM attacker 'simple' things like depending on sequence numbers might not >> work.) > > Exactly. A lot of the complexity in shim6 comes from this requirement, and > serves as a proof of concept. > >
If the payload is secured with TLS, do we need to care about the IP addresses change? The rogue endpoint shouldn't be able to extract the payload and after a while the other endpoint should just drop the subflow, since the subflow is not replying correctly. Both SCTP and MPTCP uses a 32-bit tag/token to provide a low level security mechanism, in most cases it is good enough (e.g. web browsing) - that is at least what the transport folks has concluded (my interpretation) And shim6 people have concluded that CGA or HBA must be used. Why this conflict - is it due to that IPsec is mandatory in IPv6 and similar security must be available in shim6? Or do I miss something? -- patte _______________________________________________ rrg mailing list [email protected] http://www.irtf.org/mailman/listinfo/rrg
