Has anyone created any long-term private keys using cryptlib 3.4.0?

The cryptlib 3.4.0 release introduced an improved private-key storage format
that used CMS authenticated encryption to integrity-protect the private key
data (actually cryptlib validated it on load anyway so the chances of anyone
being able to tamper with it were negligible, but the authentication provides
an extra level of checking).

Anyway, there's been a recent revision to the authenticated encryption format 
that means existing keys created with the 3.4.0 release (and only the 3.4.0 
release, not any other version) would fail their authentication check on read.

I'm trying to figure out how serious a problem this is... does anyone have
private keys created with that release that they can't easily replace?  If
not, I'll release an updated 3.4.0.1 that uses the new format.  Otherwise,
it's possible to comment out the authentication check on read to allow the
keys to be read in the old format and then written out again in the new
format, but whether it's necessary to do that depends on whether anyone has
created non-replaceable private keys using 3.4.0.

Peter.

_______________________________________________
Cryptlib mailing list
[email protected] via Mail: [email protected]
Archive: ftp://ftp.franken.de/pub/crypt/cryptlib/archives/
http://news.gmane.org/gmane.comp.encryption.cryptlib
Posts from non-subscribed addresses are blocked to prevent spam, please
subscribe in order to post messages.