Lex-2008 7 hours ago

I think having multiple authoritative DNS servers in different networks might help with this.

I'm not sure how CAs deal with situations when a domain has several authoritative DNS servers, but not all of them respond with expected reply. Would it would be a good idea for CAs to ask *all* authoritative DNS servers and require a coherent answer from them?

▲

throw0101c 7 hours ago | parent [-]

> I think having multiple authoritative DNS servers in different networks might help with this.

For the case of the dns-01 challenge type, sure. But if you're using http-01 (or the attacker asks for it), then all those DNS servers will answer back with the same IP address for the A(AAA) query and you're back to the hijacking scenario.

▲

Lex-2008 5 hours ago | parent | next [-]

> http-01

Good point, I forgot about that! But then CAA record might help, I guess, if you specify your accounturi?

▲

throw0101c 5 hours ago | parent [-]

The attacker could hijack the network and initiate the ACME protocol against the CA in the CAA field. There is a draft for dns-account-01 which allows one to specify an ACME account which is authorized for updates (which uses the account key (and not cert key)):

* https://datatracker.ietf.org/doc/html/draft-ietf-acme-dns-ac...

▲

Lex-2008 3 hours ago | parent [-]

haven't heard about dns-account-01, thanks! Heard only about dns-persist-01.

I meant to say that CAA record can also specify account, or at least Let’s Encrypt docs mention it[1]. So unless attacker has access to victim's account - they can't get a cert (unless they can hijack dns, too).

[1]: https://letsencrypt.org/docs/caa/#the-accounturi-parameter

	▲	throw0101c an hour ago \| parent [-]
		Account seems to be a later extension to CAA: * https://datatracker.ietf.org/doc/html/rfc8657

▲

codedokode 5 hours ago | parent | prev [-]

http-01 seems absolutely unreliable, any hosting provider or any government can issue fake certificates using it. What a garbage technology.

▲

mixdup 3 hours ago | parent | next [-]

The point is to prove you control a domain. If you are answering traffic to requests from clients at that domain, you control it. Whether you're supposed to have control over it is outside the scope of this validation

▲

Lex-2008 3 hours ago | parent | prev [-]

Agreed. On the other hand, these DV certificates only claim to verify who _controls_ the domain. Probably, you can agree that hosting provider and government can also control the domain, no? :)

Moreover, with dns-01, DNS provider can issue fake certificates - unless you self-host DNS - in which case, only your DNS registrar can... and a government.

	▲	throw0101c an hour ago \| parent [-]
		> Moreover, with dns-01, DNS provider can issue fake certificates - unless you self-host DNS - in which case, only your DNS registrar can... and a government. Which is what DNSSEC is supposed to help with: the registrar/government would have to replace your DS/KSK with their own. Without DNSSEC queries/replies could potentially be altered over the wire (or you're in a race condition as to whether the real or attacker's reply get back first).