Peter Eckersley <[email protected]> wrote:
    >> > Boulder does indeed follow redirects, so that's our current plan.
    >>
    >> So, if the verifier follows an HTTP redirect to an HTTPS site (does it
    >> even have to be same name?), which has a (at that point) bogus
    >> certificate, and Boulder can cope with that, when why bother having
    >> the SimpleHTTP check method at all?
    >>
    >> Why not just always use SimpleHTTPS?

    > Because it is common for webservers to be configured with HTTPS vhosts
    > that are a strict subset of the HTTP ones, plus a misallocted
    > default/wildcard HTTPS vhost that is controlled by the
    > tenant/hypothetical attacker.

Ah, so the attacker can control the wildcard HTTPS, and therefore get HTTPS
certificates for the vhosts that do not have their own HTTPS vhost.

I got it.

--
]               Never tell me the odds!                 | ipv6 mesh networks [
]   Michael Richardson, Sandelman Software Works        | network architect  [
]     [email protected]  http://www.sandelman.ca/        |   ruby on rails    [

Attachment: signature.asc
Description: PGP signature

_______________________________________________
Acme mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/acme

Reply via email to