685286 - Data-driven (not code-driven) blacklisting mechanism for certificates on issuer public key or issuer distinguished name

Status: RESOLVED FIXED

(NSS :: Libraries, defect, P1)

Description

[Brian Smith (:briansmith, :bsmith, use NEEDINFO?)]

+++ This bug was initially created as a clone of Bug #642503 +++
+++ This bug was initially created as a clone of Bug #642395 +++

Mozilla would like to have a way of blocking certificates based on a blacklist of issuer public keys and/or a list of issuer subject names, in a way that we can update via our Firefox/Thunderbird update pings, which don't require rebuilding the trusted roots module, which we can ship with the next Firefox update.

One way of enabling this: Add a new API to NSS, which allows the application to register a post-validation callback that has the opportunity to change the result and error code reported by each verification function. Then the application can implement all this logic itself.

Another way would be to implement the DN/public key blocklisting in NSS, and allow the application to register an array issuer DNs or public keys to block.

My plan is to implement the first option.

Comment 1

[Kai Engert [:KaiE:]]

Note that I filed bug 647868 which also contains detailed ideas.

Comment 2

[Phillip Hallam-Baker]

Attachment: Proposal for establishing an emergency response format based on OCSP and DNS/CAA

The attached is an outline for a proposed format for emergency response to cryptographic security incidents.

This would enable emergency revocation of EE and CA certs plus pushing out policy for the affected domains (e.g. only respect this set of CA issuers for the domain, always use HTTPS, etc.)

Comment 3

[Phillip Hallam-Baker]

While all the desktop browsers have been patched, folk in the ME are more likely to be using mobile phones. These are not patched yet:

http://news.techworld.com/security/3301828/google-and-apple-fail-to-revoke-diginotar-ssl-certificates-on-smartphone/

Comment 4

[Ian Melven :imelven]

(In reply to Phillip Hallam-Baker from comment #3)
> While all the desktop browsers have been patched, folk in the ME are more
> likely to be using mobile phones. These are not patched yet:
> 
> http://news.techworld.com/security/3301828/google-and-apple-fail-to-revoke-
> diginotar-ssl-certificates-on-smartphone/

Firefox for Android is patched also.

Comment 5

[Brian Smith (:briansmith, :bsmith, use NEEDINFO?)]

Phillip, we are not going to implement that in this bug. We already have a plan for a different implementation, which is what this bug is about. Please file a new about implementing your proposal in Product:Core Component:"Security - PSM".

Comment 6

[BugBot [:suhaib / :marco/ :calixte]]

In the process of migrating remaining bugs to the new severity system, the severity for this bug cannot be automatically determined. Please retriage this bug using the new severity system.

Comment 7

[Benjamin Beurdouche [:beurdouche]]

OneCRL