Saturday, February 19, 2005 6:55 PM bart

SHA1 also insecure?

There seem to be indications (no proof yet) that SHA1 (one of the well-known hashing algorithms, developed by NIST) is not secure (enough) anymore. According to http://www.schneier.com/blog/archives/2005/02/sha1_broken.html Wang, Yin and Yu would have found weaknesses in the algorithm. You might remember these names as these folks also discovered the weaknesses in MD5 back in August last year as well as weaknesses in MD4, RIPEMD and HAVAL-128 (paper on http://eprint.iacr.org/2004/199.pdf). However, there is no reason to panic (yet).

Nevertheless, if MD5 and SHA1 are both discovered to be insecure, functions such as FormsAuthentication.HashPasswordForStoringInConfigFile() will need a rewrite (or better, will need to be extended). Other hash algorithms that could be used are RIPEMD-160 (invented by some researchers of KU Leuven - Belgium as an answer on RIPEMD, more info on http://www.esat.kuleuven.ac.be/~bosselae/ripemd160.html) and HMAC (which in fact uses an embedded hash function and is used for things such as SSL and IPsec - RFC 2104). And sites such as http://www.secure-hash-algorithm-md5-sha-1.co.uk/ (what a nice URL) seem to be outdated for quite some time already.

Del.icio.us | Digg It | Technorati | Blinklist | Furl | reddit | DotNetKicks

Filed under:

Comments

# SHA-1 also insecure? (cont'd)

Friday, August 19, 2005 9:54 PM by TrackBack