The client and the server shouldn't both generate responses exactly the same way with the same key, no. If you use HMAC, I think including a simple identifier would be good enough. Something like this: HMAC(key, device ID + counter + timestamp), where the server and client has different IDs. Den 24 aug 2013 09:32 skrev "Jeffrey Walton" <noloa...@gmail.com>:
> Hi All, > > When a symmetric key based challenge response is used, an attacker can > perform a reflection attack by starting a second instance of a > protocol and having the server answer its own questions. > > To guard against the attack, is it sufficient to ensure all challenges > sent from server to client are equal to 1 mod 2; and all client to > server challenges are equal to 0 mod 2? Is it enough to break the > symmetry? > > Jeff > _______________________________________________ > cryptography mailing list > cryptography@randombit.net > http://lists.randombit.net/mailman/listinfo/cryptography >
_______________________________________________ cryptography mailing list cryptography@randombit.net http://lists.randombit.net/mailman/listinfo/cryptography
