To finish this, find the code below using MSCAPI to verify the ocsp sig.
Most MSCAPI function are wrapped in classes but i think you will get the
point.
HTH,
Sascha
bool OcspResponse::ValidateSignature()
{
DWORD cbDer = 0;
PBYTE pbDer = NULL;
PCCERT_CONTEXT certContext = NULL;
HCRYPTKEY certPubKey = NULL;
do {
if(m_sigvalidated)
break;
if(m_rootcert == NULL)
break;
certContext = m_rootcert->ToCertContext();
if(certContext == NULL)
break;
ALG_ID sigAlgId = Oid::GetAlgId(m_signaturealgorithm.Oid);
if(sigAlgId == 0)
break;
// gets the DER-formated RespData of the OCSP response
pbDer = m_respdata->GetBytes(cbDer);
CryptProvider cryptProvider;
cryptProvider.Load();
CryptImportPublicKeyInfo(
cryptProvider,
X509_ASN_ENCODING,
&certContext->pCertInfo->SubjectPublicKeyInfo,
&certPubKey
);
if(certPubKey == NULL)
break;
GenericHash hash(sigAlgId, &cryptProvider);
hash.Create();
hash.Update(pbDer, cbDer);
// skip the unused-bits byte
Memory::ByteBuffer sig(m_signature.ToPointer()+1,
m_signature.Size()-1);
sig.Reverse(); // reverse byte-order
m_sigvalidated = CryptVerifySignature(
hash,
sig.ToPointer(), sig.Size(),
certPubKey,
NULL,
0
) != 0;
} while(false);
Memory::SecFree(pbDer);
if(certPubKey != NULL) CryptDestroyKey(certPubKey);
if(certContext != NULL) CertFreeCertificateContext(certContext);
return m_sigvalidated;
}
> -----Original Message-----
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On Behalf Of Sascha Kiefer
> Sent: Dienstag, 1. Mai 2007 11:02
> To: [email protected]
> Subject: RE: OCSP Response Signature
>
> Maybe i figured it out:
> The first byte of a bit string tells the number of unused
> bits in the last byte of the bit string, right?
>
> > -----Original Message-----
> > From: [EMAIL PROTECTED]
> > [mailto:[EMAIL PROTECTED] On Behalf Of Sascha Kiefer
> > Sent: Dienstag, 1. Mai 2007 09:08
> > To: [email protected]
> > Subject: RE: OCSP Response Signature
> >
> > Thanks for the hint.
> > I found out that i looked at the right signature value.
> > The only difference is that my signature starts with '00'H
> (it always
> > does).
> > Looking at the actually OCSP binary response shows me that
> this '00'H
> > is really part of the data but it is somehow skipped by the openssl
> > implemenation.
> > My question now is: why is the '00'H part of the asn.1
> stream but not
> > used at all?
> >
> > Openssl signature value:
> >
> > 0000: 0d 68 18 d4 8a 35 b0 a4 - 40 79 af c3 a1 93 6f ea
> > [EMAIL PROTECTED]
> > 0010: 2a 91 51 e4 f6 07 6c a2 - 8e 52 5d 07 87 77 0e 2c
> > *æQõ÷.lóÄR].çw.,
> > 0020: bb 26 4e 07 e1 46 c5 fc - ad 56 ad b2 6a b4 12 fe
> > +&N.ßF+³¡V¡¦j¦.¦
> > 0030: c4 0d 89 3f 83 32 6f 5e - 05 eb 10 0b 21 4a ba 56
> > -.ë?â2o^.Ù..!J¦V
> > 0040: f8 53 0a 13 0d e3 57 87 - 0f e0 80 d3 a7 f9 a2 78
> > °S...ÒWç.ÓÇ˺¨óx
> > 0050: bd 40 dc 4e 95 e3 ca 0a - 4c 02 08 c0 de 5f b9 72
> > [EMAIL PROTECTED]
> > 0060: 4e b7 16 7f 10 76 89 0f - ee 59 6c 5c 48 4a cb cb
> > NÀ.¦.vë.¯Yl\HJ--
> > 0070: 5e e7 f6 82 8d b5 e0 97 - 8b 80 38 5c c7 81 a5 e5
> > ^þ÷éìÁÓùïÇ8\ÃüÑÕ
> >
> > My signature value:
> >
> > 0000: 00 0d 68 18 d4 8a 35 b0 - a4 40 79 af c3 a1 93 6f
> > [EMAIL PROTECTED]
> > 0010: ea 2a 91 51 e4 f6 07 6c - a2 8e 52 5d 07 87 77 0e
> > Û*æQõ÷.lóÄR].çw.
> > 0020: 2c bb 26 4e 07 e1 46 c5 - fc ad 56 ad b2 6a b4 12
> > ,+&N.ßF+³¡V¡¦j¦.
> > 0030: fe c4 0d 89 3f 83 32 6f - 5e 05 eb 10 0b 21 4a ba
> > ¦-.ë?â2o^.Ù..!J¦
> > 0040: 56 f8 53 0a 13 0d e3 57 - 87 0f e0 80 d3 a7 f9 a2
> > V°S...ÒWç.ÓÇ˺¨ó
> > 0050: 78 bd 40 dc 4e 95 e3 ca - 0a 4c 02 08 c0 de 5f b9
> > [EMAIL PROTECTED]
> > 0060: 72 4e b7 16 7f 10 76 89 - 0f ee 59 6c 5c 48 4a cb
> > rNÀ.¦.vë.¯Yl\HJ-
> > 0070: cb 5e e7 f6 82 8d b5 e0 - 97 8b 80 38 5c c7 81 a5
> > -^þ÷éìÁÓùïÇ8\ÃüÑ
> > 0080: e5 - Õ
> >
> >
> > > -----Original Message-----
> > > From: [EMAIL PROTECTED]
> > > [mailto:[EMAIL PROTECTED] On Behalf Of Dr.
> > Stephen Henson
> > > Sent: Freitag, 27. April 2007 20:36
> > > To: [email protected]
> > > Subject: Re: OCSP Response Signature
> > >
> > > On Fri, Apr 27, 2007, Sascha Kiefer wrote:
> > >
> > > > Thanks for your response.
> > > > Please see the other mail i wrote in response.
> > > > The one that has the complete ocsp response dump.
> > > >
> > >
> > > You should be able to use the OpenSSL ocsp utility to test
> > that stuff,
> > > including (with a debugger or printf) the expected hash
> > value of the
> > > response.
> > >
> > > Steve.
> > > --
> > > Dr Stephen N. Henson. Email, S/MIME and PGP keys: see
> > homepage OpenSSL
> > > project core developer and freelance consultant.
> > > Funding needed! Details on homepage.
> > > Homepage: http://www.drh-consultancy.demon.co.uk
> > >
> >
> ______________________________________________________________________
> > > OpenSSL Project
> > http://www.openssl.org
> > > Development Mailing List
> > [email protected]
> > > Automated List Manager
> > [EMAIL PROTECTED]
> > >
> >
> >
> ______________________________________________________________________
> > OpenSSL Project
> http://www.openssl.org
> > Development Mailing List
> [email protected]
> > Automated List Manager
> [EMAIL PROTECTED]
> >
>
> ______________________________________________________________________
> OpenSSL Project http://www.openssl.org
> Development Mailing List [email protected]
> Automated List Manager [EMAIL PROTECTED]
>
______________________________________________________________________
OpenSSL Project http://www.openssl.org
Development Mailing List [email protected]
Automated List Manager [EMAIL PROTECTED]
