Re: Question on JCE 1.5 key wraps

[Berin Lautenbach]

Sean,

test_five_content_aes128_cbc_kw_aes192
test_five_content_3des_cbc_kw_aes128
test_five_data_aes256_cbc_3des
test_five_data_aes192_cbc_aes256
Basically any interop test where a symmetric key wrap is used.  The 
wrap/unwrap tests work, which leads me to believe that I am calling 
using the wrong cipher spec for the unwrap cipher.

Unfortunately, the Sun JCE doesn't take AESWrap or DESEDEWrap, only AES 
or DESEDE, so I *think* what is happening is the JCE is doing a straight 
encrypt/decrypt rather than an AES Key Unwrap or 3DES based CMS Unwrap 
respectively.

Anyway - still researching, but all ideas welcome :>.

Cheers,
        Berin
Sean Mullan wrote:
Can you tell me which test vectors are failing?

--Sean

Berin Lautenbach wrote:

Peoples,

I have just checked in a new version of config.xml that works for most 
encryption algorithms under SunJCE (have not yet checked sig).

One question - I am having issues with symmetric key wraps.  The 
Baltimore interop tests all fail where a symmetric key wrap is used. 
The BC JCE takes an algorithm of "AESWrap" or "DESEDEWrap" for the 
unwrap algorithms.  The SunJCE , I think, uses "AES" and "DESEde" 
However I get decryption errors.

Is there something special here that someone is aware of?  (Am hoping 
on  of the Sun gurus in the list can help me here :>.)

Cheers,
    Berin