+1 - new specifications should definitely not be using MD5, and the SHA-2 
hashes are preferable to SHA-1.

A useful related reference is NIST SP 800-131A: 
http://csrc.nist.gov/publications/nistpubs/800-131A/sp800-131A.pdf

Thanks,
--David

From: nvo3 [mailto:[email protected]] On Behalf Of Truman Boyes
Sent: Friday, December 06, 2013 3:31 PM
To: Melinda Shore
Cc: [email protected]
Subject: Re: [nvo3] comments on nvo3 security requirements draft

I would recommend against suggesting MD5 as the hashing technology in a text to 
be used for future development. If necessary to describe the technology, I 
would recommend SHA-256(sha2) or SHA-512. This would not have the same issues 
of collisions as MD5 currently does.


On Thu, Dec 5, 2013 at 8:55 PM, Melinda Shore 
<[email protected]<mailto:[email protected]>> wrote:
On 12/5/13 3:50 PM, ramki Krishnan wrote:
>>>REQ2: (Page 8)
> This should recommend some authorization mechanisms such as md5 checksum.
I agree with your other suggestions, but 1) I don't think a
requirements document should be making specific technology
recommendations, and 2) md5 provides some assurances about
message integrity, but really has nothing to say about
policy.  In rereading the requirement I think it's actually
not as clear as it could be although I think its intent is
absolutely correct.  I'd probably change the text to something
along the lines of:

"Before accepting a control packet, the device receiving
the packet MUST verify that the device sending the request
is authorized to make that request.  This is a policy
decision."

Melinda

_______________________________________________
nvo3 mailing list
[email protected]<mailto:[email protected]>
https://www.ietf.org/mailman/listinfo/nvo3

_______________________________________________
nvo3 mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/nvo3

Reply via email to