"Dr. Stephen Henson" <st...@openssl.org> writes:

> On Mon, May 16, 2011, Henrik Grindal Bakken wrote:
>
>> 
>> Hi.  I'm trying to test the current CVS HEAD with
>> FIPS_set_module_mode(1).
>> 
>> It's looking fairly promising to me, but I currently have one problem:
>> While performing an SSL handshake, I get
>> 1208113320:error:060A80A3:digital envelope routines:FIPS_DIGESTINIT:disabled 
>> for fips:fips_md.c:179:
>
> The rest of OpenSSL cannnot currently use the FIPS module correctly
> in all cases. You'll get quite a few problems like this. For now
> only the things in README.FIPS will work.

Okay.

>> This sounded a bit weird to me, since I've tried my best to set up
>> my application to use only FIPS-validated algorithms, but to no
>> avail.  I added some debugging printouts to my libcrypto, and from
>> what I could understand, the digest in question is MD5.  When I
>> patched openssl to say MD5 was a FIPS-approved digest, it worked.
>> 
>> The program I'm using is attached, and also output from a separate
>> 'openssl s_client -connect -showcerts'.
>> 
>> Does anyone have any ideas as to why MD5 appears in this handshake?
>
> MD5 is a mandatory algorithm for TLS 1.1 and 1.0. As a result the
> use of MD5 is permitted solely for use in TLS in FIPS mode. Handling
> this requires some exception code in the ssl library which isn't
> currently in place for HEAD.

Aha.  I'll work around this by allowing MD5 for the time being, I
think.  Thanks for your response.


-- 
Henrik Grindal Bakken <h...@ifi.uio.no>
PGP ID: 8D436E52
Fingerprint: 131D 9590 F0CF 47EF 7963  02AF 9236 D25A 8D43 6E52
______________________________________________________________________
OpenSSL Project                                 http://www.openssl.org
Development Mailing List                       openssl-dev@openssl.org
Automated List Manager                           majord...@openssl.org

Reply via email to