Randy,

>> I'm suggesting that A delegates a unique signing key to the RS.
> 
> the expression we use is, now RS can sign gifs of naked furries in A's
> name.  i.e. A has given away the store.  you do NOT let anyone else have
> your private keys.
> 
> for example. in this context, RS can now give that key to Perp who can
> originate A's prefixes.  #fail

I do not follow this reasoning. The certificate for BGPSEC are EE certificates 
with only A's ASN in its RFC3779 extension. So, you cannot use the same key to 
sign a ROA with another ASN nor issue any certificate using that same key.

IMHO, A good idea could be to clearly identify BGPSEC EE certs in the RPKI 
repository by assign them a distinct Extended Key Usage (EKU). The use of EKU 
is permitted by the RPKI CP. The EKU should be checked by the RP during the 
validation process.

Roque  


> 
>> This is what "6.6 Proxy Signing" in
>> draft-sriram-bgpsec-design-choices suggests, is it not ?  Or does that
>> blow the trust model to hell, also ?
> 
> it does indeed.  that is why 6.6 was rejected.
> 
> randy
> _______________________________________________
> sidr mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/sidr

Attachment: smime.p7s
Description: S/MIME cryptographic signature

_______________________________________________
sidr mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/sidr

Reply via email to