Chosen-Ciphertext Attacks against PGP

[Lene Larsen]

I look at this article from 2002 http://www.schneier.com/paper-pgp.html, is
the problem about
the Chosen-Ciphertext Attacks against PGP and GnuPG still a problem. It
sounds like it is easy to decrypt a PGP and a GPG message.


Regards

Lene.

[JTF]

The page brings you to a pdf of the works. I saved the PDF and did a
properties on the file. The file was created in August 2002.
The document is here
http://www.schneier.com/paper-pgp.pdf


The attacks talked about are on PGP 2.6.2 and GnuPG version 1.x

I am sure that these flaws were addressed.

I look at this article from 2002http://www.schneier.com/paper-pgp.html, is
the problem about
 the Chosen-Ciphertext Attacks against PGP and GnuPG still a problem. It
sounds like it is easy to decrypt a PGP and a GPG message.
Regards
Lene.

[David W. Hodgins]

Post by Lene Larsen
I look at this article from 2002 http://www.schneier.com/paper-pgp.html, is
the problem about
the Chosen-Ciphertext Attacks against PGP and GnuPG still a problem. It
sounds like it is easy to decrypt a PGP and a GPG message.

That attack is based on the rfc2440 specifications, which has been superseded
by rfc4880. See section 14 of http://tools.ietf.org/html/rfc4880

Note that the attack requires the target to decrypt the known ciphertext,
and return the decrypted text to the attacker. While it is a serious concern
in some situations, in most real life usage, the attack is not worth worrying
about.

Regards, Dave Hodgins

[David E. Ross]

Post by Lene Larsen
I look at this article from 2002 http://www.schneier.com/paper-pgp.html, is
the problem about
the Chosen-Ciphertext Attacks against PGP and GnuPG still a problem. It
sounds like it is easy to decrypt a PGP and a GPG message.

This was reported to Carnegie Mellon University's CERT almost four years
ago. CERT still lists this as a "candidate" and not a confirmed
vulnerability (CAN-2005-0366).

US-CERT lists this as vulnerability VU#303094 at
<http://www.kb.cert.org/vuls/id/303094> and indicates there are
significant limitations on what an attack can accomplish.

I feel that my own analysis at
<http://www.rossde.com/PGP/pgp_weak.html#manmiddle> is still valid.
Thus, I'm not concerned at all.

[Lene Larsen]

Post by David E. Ross
This was reported to Carnegie Mellon University's CERT almost four years
ago. CERT still lists this as a "candidate" and not a confirmed
vulnerability (CAN-2005-0366).
US-CERT lists this as vulnerability VU#303094 at
<http://www.kb.cert.org/vuls/id/303094> and indicates there are
significant limitations on what an attack can accomplish.
I feel that my own analysis at
<http://www.rossde.com/PGP/pgp_weak.html#manmiddle> is still valid.
Thus, I'm not concerned at all.

Thank you, for the informative information ;o)

Lene