No, replaying a message that happens to have a DKIM signature in it is
not what we're talking about when we refer to "DKIM replay".  The
point of a DKIM replay attack is specifically to use a signature that
continues to validate in order to get false credibility.

Barry

On Wed, Jan 11, 2023 at 11:17 AM Grant Taylor
<[email protected]> wrote:
>
> On 1/11/23 6:54 AM, Dave Crocker wrote:
> > The thing that we should go out of our texts' way to make clear is that
> > DKIM Replay requires that the replay attack send mail that retains DKIM
> > /validation/.  Not the mere reuse of the signature, but the continued
> > validity of that signature, for the message being sent.  A signature
> > that fails is not a replay attack.
>
> I question that.
>
> I'd think that both are a replay.  A signature that validates would be a
> successful replay while a signature that fails is a failed replay.  But
> both are replays to me.
>
> Remember, an attack doesn't need to succeed to be considered an attack.
>
>
>
> --
> Grant. . . .
> unix || die
>
> _______________________________________________
> Ietf-dkim mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/ietf-dkim

_______________________________________________
Ietf-dkim mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/ietf-dkim

Reply via email to