>>>>> "Christoph" == Christoph Riesenberger <[EMAIL PROTECTED]> writes:
Christoph> Thanks, Tom. This means, Lowe's attack doesn't touch kerberos!? Not exactly. Attacks similar to Lowe's are effective against Kerberos version 4. Christoph> 2 other questions: Christoph> Kerberos uses symmetric keys. How can it guarantee, that a Christoph> message/ticket was not altered (integrity)? Encrypted messages in Kerberos v5 always have an integrity check. In RFC 1510, this is an encrypted plaintext checksum (which has known weaknesses). More recently, encrypted messages always use a keyed hash or HMAC for checking integrity. Christoph> How does logout work? Credentials are typically destroyed at logout time by the client system. In addition, credentials are only valid for a limited time, so a stolen credential cannot be used for an indefinite period of time. Christoph> I really looked around but found no answers. Have you read RFC 1510? Also, work is occurring to revise the specification; see draft-ietf-krb-wg-kerberos-clarifications-04.txt and draft-ietf-krb-wg-crypto-06.txt. ---Tom ________________________________________________ Kerberos mailing list [EMAIL PROTECTED] https://mailman.mit.edu/mailman/listinfo/kerberos
