RE: problem - veryfing Cyclone signed message
"Dale Moberg" <[email protected]> Wed, 17 Sep 2003 09:42:34 -0700
| Newsgroups | gmane.ietf.ediint |
|---|---|
| Message-ID | <A1C7F8CC121B444CA879FB66B3268A1AD5CF7B@usazscsmh1.cyclonecommerce.com> |
Hi Konrad, This list is primarily for discussions of the drafts from the EDIINT IETF group. Nevertheless, here is a suggestion that may help you. The "second part" is really a PKCS7 detached signature, which is a structure originally defined by the Public Key Cryptography Specifcation 7. Basically it omits the "content data part" of the signature which is usually present in a pkcs7. PKCS7 is defined using ASN.1, and encoded in BER (the certificates or CRLS that may be included will be DER encoded). Use your favorite search engine to find copies of these or just read the IETF CMS specification instead (and postpone the extensive history of readings.) I think following up on these references will eventually (and it will be a while) lead to a solution to your difficulties! Cordially, Dale Moberg -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Konrad Jasinski Sent: Wednesday, September 17, 2003 6:22 AM To: [email protected] Subject: problem - veryfing Cyclone signed message Importance: Low Hi, can anyone help me with my problem? I'm testing some self-made software to communicate with CycloneActivator ver 4.2 by using AS2 protocol; To do tests I'm using two Cyclons and my soft is analyzing backup data I'm analyzing signed messages from Cyclone, but I have a problem with verifing them. In rfcs and drafts it's said that multipart/signed message should be created with two parts: - in the first part: source data with some MIME Headers - in the second part: signature computed over the first part the first part sent by Cyclone is OK, but the second part is strange for me , it has much more that 64 bytes (Cyclone is using RSA(512 bits) with SHA-1, so the signature should be 64 bytes long) after a while i realized that in second part is: - something like certificate (several bytes are different, the public key is the same I've imported from Cyclone earlier - .p7c and .cer format) - I don't know why Cyclone is sending it... - after that some data,which loooks similar to some kind of certificate - at the end, just before "----=_Part....": 64 bytes - I thaught that it was a signature, but when Cyclone sends the same data, these 64 bytes are different (however the first part is all the time the same) I've computed hash(sha-1) over the first part and it's equal to the value returned by the second Cyclone in MDN I can't find the proper 64 bytes to verify this message, and i don't know how it's possible since these 64 bytes at the end of the second part are changing perhaps, these 64 bytes are computed over some more data (maybe including current time or sth else) i'm attaching backup files: - 2 different files with the same data sent : a106test1in , a108test1in (differences in time,----=_Part...,and those 64 bytes! ) - certificates created by sending Cyclone: cert.cer, certp7.p7c - MDN for a104test1in, returned by the second Cyclone: mdn Thanks! best regards: Konrad Jasinski