On 30/03/2017 14:11, Gervase Markham wrote:
On 28/03/17 12:21, Rob Stradling wrote:
Increased attack surface. An undisclosed dormant sub-CA most likely has
its private key in an online HSM, and so I think it's prudent to assume
that it's more vulnerable (to being compromised by an attacker, or to
being accidentally used to misissue a cert) than an offline root key.
If it's dormant, there's no particular reason the HSM will be online.
But it might be, and it doesn't make much sense to make a distinction in
the policy.
IINM, the purpose (so far) of Mozilla's intermediate cert disclosure
policy is to map the attack surface. Right?
That's certainly one goal :-)
Does a week sound about right?
As previously stated, I think this will be too short if the issuance
happens at a time when a non-CCADB root program (or the CCADB
infrastructure) is closed for holidays, such as the following:
- Christmas/New Years (Up to Dec 15 days plus the skew between Julian
and Gregorian calendars, total 28 days).
- Chinese new year (15 days plus travel, as much as 21 days).
- Easter (Up to 10 days, Saturday to Monday).
- Ramadan (up to 30 days).
For previously stated security reasons, it is problematic if a SubCA is
logged with one root program or CT log long before being logged with
another.
P.S.
I calculate Christmas as 28 days in case one root program is closed
from 23 dec to 6 Jan Gregorian, while another is closed from 23 dec to
6 Jan Julian, overlapping the other by 2 days and forcing any
coordinated disclosure to wait the full 28 days.
Enjoy
Jakob
--
Jakob Bohm, CIO, Partner, WiseMo A/S. https://www.wisemo.com
Transformervej 29, 2860 Søborg, Denmark. Direct +45 31 13 16 10
This public discussion message is non-binding and may contain errors.
WiseMo - Remote Service Management for PCs, Phones and Embedded
_______________________________________________
dev-security-policy mailing list
[email protected]
https://lists.mozilla.org/listinfo/dev-security-policy