MD5 is broken for use in digital signatures. You should not use it for that purpose. This link is to a site that demonstrates creating two different digital certificates that have the same MD5 hash. When a certification authority signs your digital certificate, it is stating that you are who you represent yourself to be. They are vouching for you. If I can get a signed certificate, then transfer the signature to another certificate, I can claim to be anyone. http://www.win.tue.nl/hashclash/TargetCollidingCertificates/
SHA-1 is weakened, but it is not possible yet to actually cause collisions, so it is still acceptable for signatures. Just be sure you move to stronger hashes as they become available in products and protocols. As for the value of signatures over simple hash checking, by encrypting the hash with your private key, you assure others that only you could have calculated the hash. Thus the trustworthiness of the hash and associated data is the trustworthiness of the holder of the private key. With unsigned hashes, the trustworthiness of the hash is the lowest trustworthiness of the combination of every person and system that handles that handles the data and the hash. If given the choice, I'd opt for a signature just about every time. Eric Lengvenis Security Architecture Wells Fargo & Co. This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose, or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation. -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of DPW 401 Sent: Monday, October 27, 2008 7:44 PM To: [email protected] Subject: Re: [FDE] Digital Signatures for checking message integrity > -----Original Message----- > On Behalf Of John jdmwood-at-gmail.com |Donald Welker| > Sent: Monday, October 27, 2008 5:51 AM > > Hi all, > > I was hoping to get some clarification about "integrity checking" with > Digital Signatures. ... > My question is this: is the level of "integrity checking" of Digital > Signatures as good as or equivalent to the plain old integrity > checking you would get if you manually compared the hashes? > > For example, are there any flaws in the digital signature verification > process which mean that it's not as good for checking integrity as if > you had the hash of the file (through some trusted manner). I hope that last isn't true, or there's no point in public key crypto. My understanding is that, for all practical purposes, a digital signature *is* a signed hash bundled with a public certificate. I seem to recall some potential weaknesses in both SHA-1 and MD5 published by Chinese mathematicians; perhaps one possible (if annoying) workaround is to use both hash algorithms? Either way, plan for SHA-3: http://en.wikipedia.org/wiki/SHA _______________________________________________ FDE mailing list [email protected] http://www.xml-dev.com/mailman/listinfo/fde _______________________________________________ FDE mailing list [email protected] http://www.xml-dev.com/mailman/listinfo/fde
