Problem encrypting to a hushmail gpg key

David Shaw dshaw at jabberwocky.com
Fri Jan 29 19:43:43 CET 2010


On Jan 17, 2010, at 12:23 PM, Sean Rima wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Hi
> 
> A friend on the pgpnet mailing list is using a hushmail.com gpg key but
> when I import it, I get:
> 
> C:\Users\Sean Rima>gpg --import < test.txt
> gpg: key C4E23A82: accepted non self-signed user ID
> ""*********@hushmail.com" <---- at hushmail.com>"
> gpg: key C4E23A82: public key ""---- at hushmail.com" <jefallen at hushmail.com>"
> imported
> gpg: Total number processed: 1
> gpg:               imported: 1  (RSA: 1)
> 
> 
> If I edit the key, I see:
> 
> pub     0s/C4E23A82  created: 2010-01-07  expires: never       usage: SC
> [ unknown] (1). "------ at hushmail.com" <------ at hushmail.com>
> 
> 
> I see there is no encrytion subkey.
> 
> If I look at the key with --list-packets, I see
> 
> C:\Users\Sean Rima>gpg --list-packets < test.txt
> :public key packet:
>        version 4, algo 3, created 1262830845, expires 0
>        unknown algorithm 3

Algorithm 3 is "RSA Sign-Only".

> :public sub key packet:
>        version 4, algo 2, created 1262830846, expires 0
>        unknown algorithm 2

Algorithm 2 is "RSA Encrypt-Only".

> :signature packet: algo 3, keyid 7853D9CDC4E23A82
>        version 4, created 1262830857, md5len 0, sigclass 0x18
>        digest algo 2, begin of digest 8b f2
>        hashed subpkt 2 len 4 (sig created 2010-01-07)
>        subpkt 16 len 8 (issuer key ID 7853D9CDC4E23A82)
>        unknown algorithm 3
> 

Both of those algorithms are deprecated in the OpenPGP spec: "RSA Encrypt-Only (2) and RSA Sign-Only are deprecated and SHOULD NOT be generated, but may be interpreted."

> I am using gpg2.0.12 (waiting for gpg4win to be compiled to latest)

The 1.4.x branch will interpret these deprecated keys (internally treating them as regular RSA with the appropriate encrypt or sign flags).  I don't think gpg2 does that.

Was this generated by Hushmail?  If so, they to stop generating keys that the spec says SHOULD NOT be generated :)

David




More information about the Gnupg-users mailing list