>> @William: The 16-bit sender ID is something we already get from protocols >> like GDOI to do IV space >> partitioning (details in https://tools.ietf.org/html/rfc6054). So the >> mistake is already there. > My memory was 8 bits, ludicrously small. Reading more carefully, they > illustrated 8 bits, but also specified 12 and 16 bits. > > But this is an opportunity to do better. Let's not repeat mistakes. > > In my own area of experience, RDMA for large storage clusters across multiple > data centers, you are going to run out quickly. Use existing IP Source header > fields to build the IV. Add and xor are sufficiently fast.
While I love the approach, there would be a significant chance for a collision in the rather small IV space. Thus, modern ciphers would break. Due to IPv6 addresses being much larger than current IVs, we have to do some explicit numbering, if we want to use the same key.
signature.asc
Description: Message signed with OpenPGP
_______________________________________________ IPsec mailing list [email protected] https://www.ietf.org/mailman/listinfo/ipsec
