[cabfpub] RFC5280

[Peter Bowen]

> On Feb 24, 2016, at 11:07 AM, Ryan Sleevi <sleevi at google.com> wrote:
> 
> 
> On Feb 24, 2016 10:56 AM, "Jeremy Rowley" <jeremy.rowley at digicert.com <mailto:jeremy.rowley at digicert.com>> wrote:
> >
> > I’ve been playing around with Peter Bowen’s certlint (an excellent tool) and, looking at the cert universe as a whole, there are some noticeable issues with the BRs and RFC 5280 that I though merited a public CAB Forum discussion.  Some of this is likely me not knowing the entire history of 5280, so I appreciated any explanation. If there’s exceptions we would like to make to RFC5280, we should probably also push a bis with IETF at the same time.
> >
> >  
> >
> 
> > 3)      Years ago, we discussed that 2047 bit certs were equivalent to 2048 bit certs (although the discussion may have occurred solely on the Mozilla mailing list).  We should codify this exception.
> 
> IMO, this is a giant hack that browsers did because CAs have trouble counting (see also: serial numbers), which itself is a statement that the underlying libraries played a very liberal definition.
> 
> I would prefer not.
> 
> 

I think there is a misunderstanding here. There has never been a requirement that the modulus contain a certain number of bits set to ‘1’.  What is required is that the modulus be a 2048-bit number.  The problem is that a 2048-bit number can have one or more of the high order bits being zero.  When calculating the modulus “size”, all an observer can do find the left-most bit set to ‘1’ and use that.  RSA moduli normally are the product of two prime numbers. OpenSSL and some other generating tools have a function that makes the top bit of each prime number to be 1 which ensures the result will have the top bit set to 1.  However a random prime could be smaller, resulting in a smaller results.

Thanks,
Peter



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cabforum.org/pipermail/public/attachments/20160224/2128a40a/attachment-0003.html>