I was pointed to a new draft:

https://tools.ietf.org/html/draft-kampati-ipsecme-ikev2-sa-ts-payloads-opt-00

It's goal is to minimize the payloads for rekeying for IKE SA's and
IPsec SA's. The use case is like 3gpp use of large amounts of IKEv2
sessions.

I think the idea is fine, but I think I would to see it differently
implemented.

I think the support notify's for this should be exchanged in IKE_AUTH,
not in CREATE_CHILD_SA, because otherwise the first rekey will run
into issues or has to use the old model, or would be asymmetric wit
initiator sending all payloads anyway but responder could omit them.

I see two options:

Still use the SA payload but use a Traffic Selector Type (eg TS_UMCHANGED)
to be able to distinguish between a mistakenly omited TSi/TSr and a
purposefully one from this doucment.

Completely change the payloads of CREATE_CHILD_SA and make it more
generally with some new payload type (eg CHILD_SA_UNCHANGED) that would
cover more things that are unchanged (like USE_TRANSPORT, COMPRESS,
TFC/padding etc). This format would be so different to also ensure it
cannot be confused with regular processing.

I would need to think a bit more about the gains and complexity of each
of these solutions. I do like the fact that it allows a rekey without
allowing to modify anything about the SA other than fresh keys.

Paul

_______________________________________________
IPsec mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/ipsec

Reply via email to