> > Since a spoofer can construct any packet they like, and NOT include any
> > authentication data, a bit in the source address seems to be the only
> > way for a receiver who cares, to know whether to drop it (because auth
> > data is missing) or accept it (because it's a legacy insecure address).
> 
> What about the receiver having two IP-addresses, one for legacy and one 
> for "secure-only" source addresses?

If the receiver indeed enforced that "secure-only" source address
can only be used with the "secure-only" destination this might work
as far as the security aspects at the IP layer.
(But I haven't thought long and hard about this to tell for sure
if this is sufficient - it is required as far as I can tell.)

But, that would imply that the receiver somehow being able to control
who uses which of its IP addresses i.e. be able to ensure that the
peers that want more secure operation get the secure-only address
and vice-versa.

Thus somehow the distinction between secure and non-secure destination
addresses need to be encoded in what is stored in the DNS (and
other places that translates "names" to IP addresses).
That seems like a fair amount of change to other parts of the system.
Do you have good ideas of how this can be done?

  Erik

--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page:                      http://playground.sun.com/ipng
FTP archive:                      ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------

Reply via email to