Re: [websec] wrt Firefox's pre-loaded HSTS list impl (was: HSTS: max-age=0 interacting with includeSubdomains)

[=JeffH <[e-mail]>]

Adam Barth wrote..
 >
 > On Tue, Aug 21, 2012 at 1:38 PM, Brian Smith <[e-mail]> wrote:
 >>> Adam Barth wrote:
 >>>> Brian Smith wrote:
 >>>>> 2. The owners of example.com decides to turn of HSTS for whatever
 >>>>> reason (perhaps the domain changed owners, or there's a compatibility
 >>>>> issue, or whatever), so they start sending out HSTS with max-age=0
 >>>>> for example.com and for all the subdomains.
 >>
 >>>> That's not a correct way of disabling HSTS after (1).  Instead, they
 >>>> need only send out an max-age=0 header for example.com itself.
 >>
 >> [...]
 >>
 >>>> They can simply initiate a request to https://example.com/ (e.g., by
 >>>> using an HTTP redirect or an HTML image element) and clear the HSTS
 >>>> state for that host name.
 >>
 >> I understand what you're saying and it makes sense. And, I agree that in a
 >> web browser that is a pretty reasonable way to handle some emergency where
 >> you have to turn off HSTS for some reason, though I think it would be quite
 >> tricky to do so in a way that is reliable.
 >>
 >> Another thing to keep in mind is that, in order to turn off HSTS, the site
 >> must comply with the browser's requirements for HSTS sites anyway;
 >> otherwise the browser will ignore your HSTS header and avoid doing the
 >> redirect or avoid loading the page with the img tag in it.

I guess I'm curious what you mean by "browser's requirements for HSTS sites" and 
why it might avoid performing a redirect ?


 >> FWIW, in Firefox we are also going to honor max-age=0 as a mechanism to
 >> disable the entries in our pre-loaded HSTS list that will ship in the
 >> browser.
 >
 > How long do you plan to cache the disable?

And can the site then turn around and issue HSTS Policy with a non-zero max-age 
to reinstate the policy?

=JeffH