Re: [tcpm] question about TCP-AO and rekeying

[Joe Touch <[e-mail]>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1



>>> If so, the entry fails. If that's what
>>> you mean by "prohibit overlaps", yes, I think we should
>>> prohibit overlaps. 
>>>
>>> If what you mean is that two MKTs with different key-ids can't overlap
>>> the same socket pair space, I don't see a problem with that.
>> That is a problem for outgoing SYNs. For those, either the connection
>> has to know a-priori which ID to use, or we need to make sure MKTs can't
>> overlap at all (ignoring keyIDs).
> 
> I'm sorry, but I don't see why. 

So let's consider outgoing SYNs.

Consider a system with two MKTs:

	MKT alpha	from ANY:ANY to JOE:80	KEYID=4

	MKT beta	from ANY:ANY to ANY:ANY	KEYID=5


So my web client wants to connect to JOE:80. The web client has not been
modified to indicate a desired KEYID; I doubt many apps will be so
modified. So I'll need to ensure that the socket pair of the SYN matches
only one MKT.

That means I can't have default keys, like beta.

So what I'm wondering is whether we:

	a) require MKTs to be unique per {socketpair,ID} tuple

	b) require MKTs to be unique per socketpair

We either need (b), or we need something else that says "first match"
and establishes an ordering to MKTs, or "best match", etc.

Joe
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAko5yAAACgkQE5f5cImnZrulywCcC09/JIAnd4B+yCLc3nYVct/5
/s4An2qFtyt9tIZGo+hpepDuFeUk8zyo
=JmIZ
-----END PGP SIGNATURE-----