Re: [lamps] RSA Key Length in draft-ietf-lamps-cmp-updates

[David von Oheimb <[e-mail]>]

On 30.07.21 17:29, Russ Housley wrote:
> draft-ietf-lamps-cmp-updates defines id-regCtrl-rsaKeyLen as:
>
>        RsaKeyLenCtrl ::= INTEGER
>
> At a minimum, I think we need to make this a positive integer.
>
> I see that other groups are pushing for bigger minimum key sizes for RSA and D-H.
>
> What do others think?

I don't think it's worth the hassle making restrictions on key lengths 
at this point.

On 30.07.21 18:21, Salz, Rich wrote:
>>    Of course, a one bit RSA key is silly.
> Interesting philosophical question, I suppose. Is 1 a prime?

None of the usual key lengths (including RSA modulus size) is a prime.


On 30.07.21 17:32, Salz, Rich wrote:
>>     I see that other groups are pushing for bigger minimum key sizes for RSA and D-H.
> If we can avoid saying anything about this, that would probably be easiest.

I agree here.

In my view,

       RsaKeyLenCtrl ::= INTEGER

is entirely sufficient for the syntax of RSA key lengths.

It is understood that only sensible values should be used there,
and I see little point in trying to force users to do this at syntactic 
level.

     David