[cabfcert_policy] Entropy in Certificate Serial Numbers

[Mads Egil Henriksveen]

Hi Ben

This means that we are changing the entropy requirement from 20 bits to 64 bits. This is ok for me, but is this done intentionally?

And I understand that the entropy is not required for root certificates. This is also fine, but I would like to understand why this is required for other certificates signed by an (offline) Root CA.  To me there is a major distinction between certificates signed by an offline Root CA and those signed by an online CA.

Regards
Mads

From: policyreview-bounces at cabforum.org [mailto:policyreview-bounces at cabforum.org] On Behalf Of Robin Alden
Sent: 17. februar 2016 23:43
To: 'Ben Wilson'; policyreview at cabforum.org
Subject: Re: [cabfcert_policy] Entropy in Certificate Serial Numbers

Hi Ben,
              I'm fine with the 'unpredictable bits' part, but the serial number thing is already covered in RFC5280.
Why do we need it again in the BRs?

https://tools.ietf.org/html/rfc5280#section-4.1.2.2
says..

"The serial number MUST be a positive integer assigned by the CA to each certificate.  ..."

Robin


From: policyreview-bounces at cabforum.org<mailto:policyreview-bounces at cabforum.org> [mailto:policyreview-bounces at cabforum.org] On Behalf Of Ben Wilson
Sent: 17 February 2016 11:46
To: policyreview at cabforum.org<mailto:policyreview at cabforum.org>
Subject: [cabfcert_policy] Entropy in Certificate Serial Numbers

What about  this version of a proposed revision to Section 7.1 of the  BRs?

For all Certificates issued after _______, serialNumbers MUST be greater than zero  (0), and for Certificates issued to Subscribers and Intermediate CAs, the serialNumber MUST contain at least 64 unpredictable bits.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://cabforum.org/pipermail/policyreview/attachments/20160218/ab0717e3/attachment.html