Re: Sub/Not Security, Presence service identifiers

"Mark Day" <[email protected]> writes:

> I guess I'm a little confused, then. In a previous message, I think with
> respect to the same example, you wrote
> 
> "Signing the presence information by the person that created it does
> not fix the problem, because a spoofing presence service could collect
> signed presence information and replay them in an order that
> misinforms the watcher."
> 
> I don't understand how this style of replay attack is viable if the presence
> information signed by the creating principal includes a timestamp covered by
> the signature.

I'm afraid I don't know where I made this comment, however, the
concern is that a spoofing presence service will give a watcher old
information, even when updates are available.  Since the content
covered by the signature does not say when the notification operation
was performed by presence service, a spoofing presence service can lie
about the timeliness of the information, and the watch will never be
the wiser for it.

> Could you clarify whether you are concerned about a replay attack or
> some other form of attack, and which parts of the system have to be
> subverted for that attack to succeed?  It would be helpful to me, at
> least, for you to be very explicit about these points because some
> things that may be obvious to you are not obvious to me.

I hope I have answered these questions in other posts today.  I
attempted to clarify the example that demonstrates the value of having
a presence service identifier.  If I have not answered your questions,
please ask again.

John



  [reminder: [email protected] for non-technical discussions, please]