[DUG] Sending a file and verifying the data

[John Bird]

Looks like MD5 hashes are deprecated now....there has been security papers 
about possible generation of any MD5 hashed data using large 
computation.....(they used 200 networked PS3's if I recall) sometime around 
Xmas.

It caused a bit of a scare in the browser communities (IE/Firefox etc) as 
some of the SSL certificate authorities such as Comodo or a subsidiary 
thereof rely on MD5, although most have now switched to using SHA hashes. 
The worry was that while some recognised certificate vendors were still 
using MD5 there was the posssibility they could validate any site 
certificate even if they were using other hashes by supplying a valid MD5 
verifification I understand.

see

http://www.heise-online.co.uk/security/25C3-MD5-collisions-crack-CA-certificate--/news/112327


"The infrastructure of Certification Authorities is meant to prevent this 
kind of attack, but despite warnings, some root CAs are still using MD5, 
leaving people potentially exposed to the possibility of forged 
certificates. The team found the following CAs still using MD5; RapidSSL, 
FreeSSL, TC TrustCenter AG, RSA Data Security, Thawte and verisign.co.jp. 
They collected 30,000 certificates and found 9,000 of them were signed with 
MD5 and of them, 97 per cent were issued by RapidSSL. Because of this and 
other attributes of RapidSSL's procedures, such as use of sequential serial 
numbers in issued certificates, the researchers examined RapidSSL's 
certificates in greater depth.

By purchasing a certificate and then getting it reissued a number of times, 
data allowing prediction of the serial number was obtained, allowing the 
researchers to generate the certificate data to be signed over the course of 
just a few days. The predicted serial number was then passed to the 
Playstation 3 cluster which was asked to calculate both legitimate 
certificate data and bogus certificate data, which when MD5 hashed, would 
collide. When it came to the time the predicted serial number would be used 
by the CA, the researchers purchased a new legitimate certificate, hoping to 
get a certificate with the same serial number as they had predicted. It took 
four attempts to get the methodology to work and actually get a certificate 
with the same serial number, but the signature of the issued certificate was 
now valid on the bogus colliding certificate because of the MD5 collision."

I understand RapidSSL hurriedly switched in January...

I presume this means for Delphi its a good idea to use something 
else.....what do others use?

John

> This popped up on DelphiFeeds.com today
> http://delphi.about.com/od/objectpascalide/a/delphi-md5-hash.htm
>