Michael Ströder wrote:
Eddy Nigg wrote:
On 12/05/2008 08:58 AM, Ian G:
> When I lose a key, all the old encrypted email is no
longer readable ... which presumably happens when revocation happens as
well.
For your protection, yes.
???
If the private key is no longer available, yes, encrypted data
technically cannot be decrypted anymore.
Note the decision here to store the email in private-key encrypted form,
instead of (for example) cleartext or re-encrypting it with the master
password.
This raises a lot of questions; why this implicit choice makes sense to
a user, for example, and what happens when a key is lost: will the user
ever return to using crypto ever again?
If the public-key certificate was revoked but the accompanying private
key is still available you should be able to decrypt archived S/MIME
messages just like when the public-key certificate expired.
But a sender MUST not use your public-key certificate anymore for
generating a new encrypted S/MIME message to you and you MUST NOT use it
for generating a new digital signature anymore. Just like when the
public-key certificate expired.
If NSS is doing it differently I'd consider this a bug. I'd appreciate
if NSS developers could shed some light on this. If in doubt this should
be clarified (e.g. on ietf-smime mailing list).
Michael, just to clarify; the key that was "lost" was unable to export
from a Tbird to another. Now, it was a bit more complicated and I
couldn't figure it out; I'm not so worried about it.
I then started thinking about revocation, as the official way to "lose"
the key, and wondered if revocation would kill the encrypted email. But
I wasn't able to figure out -- quickly -- just how Tbird deals with
revocation.
All of which is meant to indicate that I haven't clarified the bug to
file as yet; it is worryingly brittle, but it may be doing the right
thing and I may be doing the wrong thing.
iang
_______________________________________________
dev-tech-crypto mailing list
dev-tech-crypto@lists.mozilla.org
https://lists.mozilla.org/listinfo/dev-tech-crypto
