On Thu, Feb 6, 2020 at 12:09 PM Eric Rescorla <[email protected]> wrote:

>
> Second, the text in question is about the effect of attacks on DNS on the
> Web "Users may be directed to bogus IP addresses for e.g. websites where
> they might reveal personal information to attackers."
>

I agree that the WebPKI can help here, but it cannot be viewed as a solid
remediation. For one, browsers will still connect to non-HTTPS sites. Here
is a large concrete example of an attack the matches what the draft
describes:

https://www.zdnet.com/article/brazil-is-at-the-forefront-of-a-new-type-of-router-attack/

Secondly, root certificate stores are not consistent. One example:

https://www.cs.umd.edu/class/fall2017/cmsc818O/papers/tangled-mass.pdf

Thirdly, it's not clear that misrouted traffic (either via DNS or lower
level ways like BGP) won't encounter seemingly-legitimate certificates,
because the number of people that can issue them is so large.

I know Ekr knows all of this, and I do think that text on bogus IP
addresses needs a caveat. But I'm not sure the text in the draft is totally
misguided.

thanks,
Rob
_______________________________________________
dns-privacy mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dns-privacy

Reply via email to