Hi Bernard, >So > adding 1.5 round-trips for an initial exchange will typically not > represent > much of a performance penalty in intranet scenarios.
Does the server send no empty certificate_request in each TLS session? If then it will cost at least 2 more round-trips. Whatever, my concern is about charging the server, especially when forged clients try to connect. The authentication server will not be able to detect such an action early. >the set of Identity protection ciphersuites described in the > document are limited, so that an EAP-TLS implementation might not be able > to > negotiate the ciphersuites that it would prefer along with identity > privacy. Recently, there was a discussion on the TLS mailing list* regarding static DH implementation, showed that the static DH ciphersuites rarely seems to be supported, especially with (http://www1.ietf.org/mail-archive/web/tls/current/msg00856.html) - The Certicom SSL-C. - cryptlib (definitely doesn't do static DH). - PureTLS (hasn't tested static DH). - OpenSSL I don't know if TLS implementations of Microsoft support static DH ciphersuites. Identity protection ciphersuites supports all ciphersuites except those based on static DH (static DH may be added, but not a full identity protection will be therefore provided). Best regards, Badra _______________________________________________ Emu mailing list [email protected] https://www1.ietf.org/mailman/listinfo/emu
