On 08/04/15 01:31, Richard Barnes wrote:
> decides to require of them.  The purpose of this thread is to discuss what
> additional steps, if any, we should require.
> 
> CNNIC has already provided Mozilla with a list of certificates issued
> before 1 Apr 2015.  We are working on publishing this list.  CNNIC has also
> informed Mozilla that they plan to take the following steps:

> D. Require CNNIC to implement Certificate Transparency [2][3]

The Google blogpost that you reference says "CNNIC will implement
Certificate Transparency for all of their certificates prior to any
request for reinclusion." So, assuming CNNIC are also interested in
being re-enabled in Chrome, it seems like there is no need for Mozilla
to require this - CNNIC will be doing it anyway. Every certificate they
issue will be publicly logged.

> E. Require a certain amount of time to pass before CNNIC's re-inclusion
> request will be considered.

I think there is value in this. If CNNIC know that a reinclusion request
will not be considered for at least time period X, that removes from
them the temptation to rush the work they need to do in order to be
ready as soon as possible.

> F. Require the re-enabled CNNIC roots to have name constraints that
> restrict them to TLDs for which they are authoritative.

As far as I can see, CNNIC is the registry for five TLDs:

..cn
..xn--fiqz9s (.中國; .china in traditional characters)
..xn--fiqs8s (.中国; .china in simplified characters)
..xn--io0a7i (.网络; basically .net in Chinese)
..xn--55qx5d (.公司; basically .com in Chinese)

https://www.iana.org/domains/root/db/cn.html
https://www.iana.org/domains/root/db/xn--fiqz9s.html
https://www.iana.org/domains/root/db/xn--fiqs8s.html
https://www.iana.org/domains/root/db/xn--io0a7i.html
https://www.iana.org/domains/root/db/xn--55qx5d.html

If we take the position that CNNIC is effectively a government CA, then
that would suggest (if we take this route) restricting them to the first
three on that list. However, it makes no sense to prevent a CA issuing
certs for TLDs for which it is also the organization which controls the
DNS, given the many ways of validating cert issuance which directly or
indirectly use the DNS. So that would suggest restricting them to all five.

Gerv
_______________________________________________
dev-security-policy mailing list
dev-security-policy@lists.mozilla.org
https://lists.mozilla.org/listinfo/dev-security-policy

Reply via email to