Tim
18/04/07, 23:23
> I appreciate you replying, but I understand the Windows DNS attack well.
> I'm just wondering how and if BIND protects against the same attack, and
> if yes, how?
Well, as the main vulnerability implies, a sane DNS cache wouldn't
accept a record that wasn't requested. If I ask for A, and I get A and
B back, and B isn't reasonably related to A, ignore B.
I'm not saying BIND is sane, but from what I understand, in this case
they got it right. The birthday attack is merely another vector to
exploit the real problem.
tim
> I'm just wondering how and if BIND protects against the same attack, and
> if yes, how?
Well, as the main vulnerability implies, a sane DNS cache wouldn't
accept a record that wasn't requested. If I ask for A, and I get A and
B back, and B isn't reasonably related to A, ignore B.
I'm not saying BIND is sane, but from what I understand, in this case
they got it right. The birthday attack is merely another vector to
exploit the real problem.
tim