[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: delta-CRLs (was Re: Last Call:draft-ietf-pkix-new-part1-06.txt comments)



Title: RE: delta-CRLs (was Re: Last Call:draft-ietf-pkix-new-part1-06.txt comments)
Sharon,
 
Thanks for shining some light on the X.509 delta CRL discussions.  So it actually IS
the intent of X.509 to permit a dCRL to reference multiple base CRLs?  That was the
interpretation that I placed on the words from section 9  X.509 Ed. 4 draft v6:
 "A dCRL may also be an indirect CRL in that it may contain updated revocation
information related to base CRLs issued by one or more than one authorities."
I was beginning to think I had misinterpreted the text. 
 
Let me see if I can figure out how these multiple base CRLs are referenced in a dCRL.
 
From X.509 Section 9 : "A dCRL includes either a deltaCRLIndicator or a crlScope
extension to indicate the base revocation information that is being updated with this dCRL."
 
But the deltaCRLIndicator extension references the base CRL issuer only by implication,
i.e. it is assumed that the base CRL issuer is the same as the dCRL issuer.
 
So if a dCRL references multiple base CRLs, it must do so via the baseRevocationInfo
component of the crlScope extension. 
 
From X.509 Section 9 :  "Because of the potential for conflicting information a CRL shall
not contain both the deltaCRLIndicator extension and a crlScope extension with the
baseRevocationInfo component."
  
So a dCRL that references multiple base CRLs cannot include the deltaCRLIndicator extension.
 
      -  Sharon, is that right?
 
I don't see anywhere in draft-ietf-pkix-new-part1-05 where it says that a delta CRL must be
identified by a deltaCRLIndicator extension.  It does say that "The delta CRL indicator is
a critical CRL extension that identifies a CRL as being a delta CRL."  This appears to be
a sufficient condition, but not a necessary condition, for the CRL to be interpreted as a 
delta CRL.  So a PKIX-profile-compliant CRL-issuer could use the crlScope extension rather
than the deltaCRLIndicator extension to identify delta CRLs?
 
      -  Russ, is that right?
 
From draft-ietf-pkix-new-part1-05 section 5.2.4: "Both the delta CRL and the complete CRL
MUST
include the CRL number extension
(see sec. 5.2.3). The CRL number extension in the
delta CRL and the complete CRL MUST contain the same value. "
 
But in order for the CRL number extension in the delta CRL and in the complete CRL to
contain the same value, there would have to be only one base CRL per delta CRL or
there would have to be multiple CRL number extensions in the delta CRL (without a good
way of matching them to the base CRLs).
 
I guess draft-ietf-pkix-new-part1-05 isn't intended to support multiple base CRLs per delta CRL.
 
      -  Russ, is that right?
 
Just musing  ......  

Regards,

Carlin

____________________________

-  Carlin Covey
   Cylink Corporation

 

-----Original Message-----
From: Sharon Boeyen [mailto:sharon.boeyen@xxxxxxxxxxx]
Sent: Monday, April 23, 2001 1:10 PM
To: 'Carlin Covey'; Russ Housley; ietf-pkix@xxxxxxx
Subject: RE: delta-CRLs (was Re: Last Call:draft-ietf-pkix-new-part1-06.txt comments)

Carlin, you are correct. That 'indirect delta' is a new facility added into the 4th edition X.509. One example that was used in those discussions was where there might be several CAs operating within a given domain (perhaps a large multinational organization) and each issues its own CRLs on a regular basis. Once retrieved, these CRLs may be cached at the validation step and used until their expiry. However, that multinational may want to issue a single delta CRL every minute and that CRL would contain revocation notices for all the CAs within the organization but perhaps only for the keyCompromise reason. I think you raise an interesting point because in this type of environment the CA that issues the delta could not issue any single corresponding base. However, I suspect indirectDeltas are beyond the scope of what 2459 plans to cover?

Sharon

>