Windows IT Pro is the leading independent community for IT professionals deploying Microsoft Windows server and client applications and technologies.
  
  
  Advanced Search 


December 21, 2004

Discovering 44 Security Holes Doesn't Make the Grade

RSS
Subscribe to Windows IT Pro | See More Security Articles Here | Reprints | Or get the Monthly Online Pass—only $5.95 a month!
back to blog index

What if you were taking a computer science course and a primary requirement is that you must discover 10 new security holes or you won't get a passing grade. Could you do it?  Hard to say, right? 

The requirement is reality for a body of students taking
D.J. Bernstein 's class entitled "MCS 494: Unix Security Holes."

According to a post on the Bugtraq mailing list (thanks to Thor Larholm at PivX Solutions) there are 25 students in the class and all totalled they have discovered
44 new security holes in Unix platforms. Doing the math, that's about 206 discoveries short if everyone were to meet the passing requirements. At least one student said "After 300 hours of work and an A average on the exams, I expect to fail the course."

Nevertheless, finding 44 holes in Unix is quite an accomplishment and it does lead to more secure operating systems. Still somehow I find it troubling that the professor has made the assumption that there are 250 holes in Unix platforms for his students to discover.

How could he expect that to be true? It might be true, but what if it isn't? Then students fail the course! And if that turned out to be the case then who would have really failed?

End of Article



Reader Comments
Doesn't look like the 25 students are required to come up with unique results.

If that is so then just 10 holes are theoretically enough to pass all students.

Anonymous User December 22, 2004


To the previous poster:

Not so. Read the item at Slashdot:

http://it.slashdot.org/article.pl?sid=04/12/15/2113202

Anonymous User December 22, 2004


To the previous poster:

I read the item at Slashdot and it doesn't say they have to be unique among all classmates. I mean, in order for them to be unique, everyone would have to tell everyone else what they found. What happens if there is a tie? Or two people are testing the same one and one person simply submits it without testing as deeply as the second? All this really proves is that there are STILL plenty of holes in *nix OSes, even after all these years of open sores examination of the code... i mean, to hear the linux advocates put it, by now all this code should have been debugged 110%

Anonymous User December 31, 2004


You must be a registered user or online subscriber to comment on this article. Please log on before posting a comment. Are you a new visitor? Register now





Search Security Matters
 
Security Matters
FEBRUARY 2010
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28       
or

 Recently in Security Matters
Wordpress 2.8.4 Fixes A Big Security Hole
Make a Comment
Microsoft Releases 5 Critical Patches
Make a Comment
How To Help Secure HTTP Data Without SSL

Last Comment
The article has very less information. Need some elaboration....
(1 Comments)
Sometimes The Cookie Doesn't Crumble
Make a Comment
SecureTweets for Twitter
Make a Comment

More blogs about technology,
software, and Windows.

Windows IT Pro Home Register FAQ for Windows WinInfo News
Europe Edition About Us Contact Us/Customer Service Media Kit Affiliates / Licensing  
SQL Server Magazine Office & SharePoint Pro DevProConnections IT Job Hound
Left-Brain.com Technology Resource Directory asp.netPRO ITTV Windows SuperSite 
 
 Windows IT Pro is a Division of Penton Media Inc.
 © 2010 Penton Media, Inc. Terms of Use | Privacy Statement