Ben-san,

Regarding Policy 2.8: MRSP Issue #178: Sunset SHA1, we agree with the response 
with a view to improving the reliability of network security.

On the other hand, the CRL signed with SHA-1 of a Root CA Certificate (Security 
Communication Root CA1) (* 1) is used for the verification of the already 
issued client-auth certificate, and suppose if we try replace the verification 
logic to SHA-2, the serious impact on the PKI of client-auth may be expected, 
which makes difficulty doing it.  For that reason, we are planning to use the 
SHA-1 CRL until September 2023, which is the expiration date of Security 
Communication Root CA1. Therefore, in order to comply sunset of the SHA-1 CRL 
before that date, we believe that it is necessary to erase the security bit of 
server-auth and revoke the Cross root certificate.
We are currently working on the timeline planning for these procedures. We also 
understand that this validation path is used by the feature phones (old cell 
phones) with a hard-coded trust store.  (As reported in # 8.d of April 2021 CA 
Communication (* 2), NTT DoCoMo, a Japanese mobile phone top carrier, plans to 
continue providing services to those feature phones until March 2026). We 
suppose that users who still use these feature phones include a significant 
proportion of elderly people, and that they will pay greater switching costs 
than ordinary consumers would do. In such a situation, we also had received 
requests from some stakeholders to let them have a certain period of informing 
time before revoking the Cross-Root Certificate, so that we are trying to 
decide timeline bit more carefully compare to other cases.

(*1)  SHA1 Root CA Certificate (Security Communication RootCA1)
https://crt.sh/?id=144
Sep 30 04:20:49 2023 GMT

(*2)  #8.d of April 2021 CA Communication
https://ccadb-public.secure.force.com/mozillacommunications/CACommResponsesOnlyReport?CommunicationId=a054o00000EL1Fo&QuestionId=Q00156,Q00151,Q00158

Thank you for your consideration.

Best regards,
Hisashi Kamo


From: 'Bruce Morton' via [email protected] 
[mailto:[email protected]]
Sent: Friday, February 4, 2022 6:13 AM
To: [email protected]
Cc: [email protected] <[email protected]>; [email protected] 
<[email protected]>; [email protected] 
<[email protected]>
Subject: Re: Policy 2.8: MRSP Issue #178: Sunset SHA1

Entrust has already sunset SHA-1 for S/MIME certificates. Entrust will respond 
to ballot SC53 and will stop using SHA-1 with OCSP responses and will push this 
requirement to CRLs as well. We have also reached out to our subordinate CAs to 
advise of the pending change and confirm their current SHA-1 status.
On Thursday, February 3, 2022 at 11:28:33 AM UTC-5 
[email protected]<mailto:[email protected]> wrote:
Hello,

This is Cybertrust Japan. One of our root CAs uses SHA-1 for CRL signing.
But we would like to sunset the use of SHA1.  In fact, our plan is to retire 
this SHA-1 Root of SecureSign Root11 and replace it with its successors. So we 
are preparing root inclusion requests.


Best,
Mo
2022年2月3日木曜日 2:35:58 UTC+9 [email protected]:
For the sake of completeness: Let's Encrypt / ISRG does not sign SHA-1 hashes 
for any purpose, and would be amenable to any sunset date.

We do accept signatures over SHA-1 hashes of CSRs provided by subscribers, and 
of course accept SHA-1 hashes for the issuerKeyHash and issuerNameHash in OCSP 
requests, but those are not relevant to this proposal.

Aaron
On Tuesday, February 1, 2022 at 7:59:37 PM UTC-8 
[email protected]<mailto:[email protected]> wrote:
I have emailed CAs in the Mozilla program asking them to respond here.

On Wed, Jan 26, 2022 at 12:41 PM Ryan Sleevi 
<[email protected]<mailto:[email protected]>> wrote:


On Wed, Jan 26, 2022 at 2:00 PM Ben Wilson 
<[email protected]<mailto:[email protected]>> wrote:
See responses inline below.

On Tue, Jan 25, 2022 at 11:12 PM Ryan Sleevi 
<[email protected]<mailto:[email protected]>> wrote:
It’s not clear: what situations make it appropriate for a CA communication, 
versus discussion here?

Yes.  It is preferable that discussion take place here. However, a survey would 
still be public, as they have been in the past, and the CCADB would collect all 
of the responses in a table format.

Oh, for sure :) I just know that the surveys have historically had delays or 
had confusion by CAs in interpreting questions, and the survey approach 
somewhat predates the m.d.s.p. participation requirement. I totally realize 
that it has benefits for bringing direct awareness, but I raise it to try and 
understand if the expectation is to always have the two parallel paths for 
soliciting feedback, or if it might just be sufficient to email blast CAs to 
say "Hey, here's the discussion, to send feedback, please participate here". 
That, I think, might achieve the goal of highlighting the importance, while 
still centralizing some of the conversation :) Just a thought

--
You received this message because you are subscribed to the Google Groups 
"[email protected]<mailto:[email protected]>" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to 
[email protected]<mailto:[email protected]>.
To view this discussion on the web visit 
https://groups.google.com/a/mozilla.org/d/msgid/dev-security-policy/b75895bc-37e7-4962-afdb-8841dd8b39c2n%40mozilla.org<https://groups.google.com/a/mozilla.org/d/msgid/dev-security-policy/b75895bc-37e7-4962-afdb-8841dd8b39c2n%40mozilla.org?utm_medium=email&utm_source=footer>.

-- 
You received this message because you are subscribed to the Google Groups 
"[email protected]" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/a/mozilla.org/d/msgid/dev-security-policy/0bdca4214f974f91b93d27b8af02ee93%40secom.co.jp.

Reply via email to