RE: Online Certificate Revocation Protocol

[Santosh Chokhani <[e-mail]>]

Marc:

I fail to see a security risk in a key that has been revoked.  Absent
trusted time stamp, all the unnecessary revocation does is cause denial of
service.

If the PKI always was implemented in conjunction with trusted time stamp,
standards fully address the issue of validation (e.g., X.509 and PKIX path
development) and the products incorporated all this, it might work.  Even
then, there is an issue of CRL size growth.

There is little benefit in revoking just in case type scenario.  That way, a
key that is not destroyed and is being used is more suseptible to
compromise.  So, if we just kept on rekeying and revoking old certificates,
it will be a pretty poor solution.

-----Original Message-----
From: Marc Branchaud [mailto:[e-mail]]
Sent: Monday, June 11, 2001 4:39 PM
To: [e-mail]
Subject: Re: Online Certificate Revocation Protocol




Two replies in one...
 
Santosh Chokhani wrote:
> 
> Revocation of a public key certificate whose companion key has been
> destroyed is a BAD idea.
>  
> For example, if the subject of the key is a CA, revocation of that
> public key certificate could cause denial of service for all the
> certificates issued by that CA.  There is nothing wrong with the
> certificates.

Just because the revocation mechanisms aren't up to the task is a poor
reason
to not take the precaution.


Santosh Chokhani wrote:
> 
> Again, it is trust issue.  I have a very simple point.  If you trust
> the holder of private key, you do NOT revoke a certificate.  If you
> do not trust the holder of private key, you probably want to do
> something whether the key was destroyed or not.

Trusting the holder is not enough.  The holder may consider the key
destroyed
because it's beyond his ability to recover, but that doesn't mean that it
can't be recovered by someone else.  Such recovery is less likely to be
noticed with an unused key.

The safe course is to revoke.  To do otherwise has security implications.

		Marc