FYI

 

Lily Chen from NIST and I co-authored a paper on the (in)security of
tunneled authentications, including  tunnel-based EAP methods.

Unfortunately, it seems to be too late to reference the analysis in the
tunnel requirement draft, but I hope that some people still might find
it interesting.

 

"An Inconvenient Truth about Tunneled Authentications"

Katrin Hoeper and Lily Chen,

to appear in the proceedings of 35th Annual IEEE Conference on Local
Computer Networks (LCN 2010)

http://techpubs.motorola.com/IPCOM/187779

 

ABSTRACT

In recent years, it has been a common practice to execute legacy
authentication protocols inside a protective tunnel. Soon after their
introduction, man-in-the-middle (MitM) attacks on tunneled
authentications were revealed and cryptographic bindings have been
applied as a countermeasure. In this paper, we demonstrate that tunneled
password-based and other types of authentication methods are still
susceptible to MitM attacks despite the use of cryptographic bindings or
other proof of binding methods. Here, so-called protective tunnels do
not protect from all attacks, and even worse yet, give users a false
sense of security. In fact, cryptographic bindings can only thwart the
attacks if the tunneled method provides strong authentication and strong
key establishment. However, such methods can be securely executed
without a tunnel. Our analysis shows that there can be no 'universal"
countermeasure because the effectiveness of a proof of binding method
depends on the properties of the authentication protocol executed inside
the tunnel. This result is unsettling, because commonly used tunneled
authentication methods, such as EAP-FAST and PEAP, do just that, i.e.,
apply one countermeasure (cryptographic bindings) and allow any type of
authentication protocol to be executed inside the tunnel. As additional
results, we show that 1) the secure derivation of traffic protection
keys depends on the type of tunneled authentication method and the
applied MitM countermeasures, and 2) security policies intended to
thwart attacks depend on the configuration of the client device and are
not practical in many environments.

 

Regards,

Katrin

 

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

Reply via email to