RE: comments on draft-ietf-smime-examples-08.txt
"Blake Ramsdell" <[email protected]> Thu, 31 Oct 2002 03:14:00 -0800
| Newsgroups | gmane.ietf.smime-examples |
|---|---|
| Message-ID | <!~!UENERkVCMDkAAQACAAAAAAAAAAAAAAAAABgAAAAAAAAARMPfbnbp50SwK3EZjypY2MKAAAAQAAAArd3QETLZTEOt/tYpYsrm/[email protected]> |
> -----Original Message----- > From: [email protected] > [mailto:[email protected]] On Behalf Of > Yves Duhem > Sent: Tuesday, October 01, 2002 7:56 AM > To: ietf-smime-examples > Subject: comments on draft-ietf-smime-examples-08.txt > > I have have done a few tests on the CMS S/MIME examples and found > some problems : > -the old OID "1.3.14.3.2.29" for SHA1withRSA is used in Alice's RSA > certificate (AliceRSASignByCarl.cer) instead of > "1.2.840.113549.1.1.5" > (which is the one given in the ASN.1 description). Agree. This is significant and must be fixed. > -the SubjectKeyIdentifier in the example 5.7 is 20 octets long while > in the corresponding certificate the extension's value is 22 > octets long > (the first two octets are missing). I believe that you have incorrectly extracted the SubjectKeyIdentifier from the certificate. The SubjectKeyIdentifier in the certificate is 20 bytes long -- please remember that you must remove a second layer of OCTET STRING to get to the juicy center (the extension value is encoded as an OCTET STRING, and the KeyIdentifier type is also an OCTET STRING). > -the S/MIME examples have a <CR> end of line whereas it > should have been > <CR><LF> An updated 5.8.eml and 5.9.eml have been created, but I believe that this is an outstanding issue for 6.8.eml. This is not a cryptographic change, so just changing the EOL characters to CRLF is sufficient. Side note (Paul) -- the timezone is a bit nutty (-0360) in 6.8.eml. > -the 5.9 S/MIME example have a protocol parameter in its Content-Type > header with a value of application/pkcs7-signature instead of > "application/pkcs7-signature" (and this parameter does not > appear in the > description of the example, only in the base64) 5.9.eml is in the process of being fixed. > -the descriptions often do not match the examples. This is being worked on also. Thanks for the comments... Blake