Dick Hardt <[EMAIL PROTECTED]> writes:

> On 19-Jun-06, at 2:59 PM, Eric Rescorla wrote:
>
>> CARRIED NONCRYPTOGRAPHICALLY INSIDE HTTP
>> One approach that comes up quite a bit is to just embed
>> authentication tokens (typically generated by an IdP) in
>> the HTTP messages that go from the user to the RP (DIX is
>> an example of such a scheme). In this system, the tokens
>> themselves are "bearer"--they enable any holder to impersonate
>> the user. In order to protect against token capture, the tokens
>> are generally made specific to the RP, so that one RP can't
>> use them to impersonate the user to another.
>>
>> An additional issue is hijacking: without cryptographic
>> binding, an attacker who observes a request can make future
>> requests in the name of the user via a cut-and-paste attack
>> (remember, the token is bearer). If you take the threat of
>> this type of attack seriously, you need to run the traffic
>> over SSL/TLS. [1]
>>
>> The primary attraction of this kind of design is that it
>> may be implementable without changing the client software
>> at all (again, see DIX). Many people see that as important
>> for deployment.
>
> In DIX, the RP includes a nonce in the request, which must then also
> be in the nonce which would prevent replay attacks assuming the RP is
> managing nonce state would it not?

Only if each authentication token is only single-use. Otherwise,
an attacker can replay it during the validity period. Even then,
cut-and-paste attacks are still possible if you block the
initial request.


> I saw the security risk here being the reliance on DNS for identity
> of the IdP in the verification step.

Hmm.... I think this depends on the design. If you're using
SSL/TLS, you should be able to block most attacks of this
class, provided you have a CRA authentication method...

-Ekr




_______________________________________________
dix mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/dix

Reply via email to