Re: OT? Are chroots immune to buffer overflows?

From: Jason Haar (Jason.Haarat_private)
Date: Wed May 22 2002 - 14:05:58 PDT

  • Next message: aazubel: "Re: OT? Are chroots immune to buffer overflows?"

    On Wed, May 22, 2002 at 08:00:25PM +0200, lorenzo wrote:
    > as others have already stated, no, you can execute anything from an
    > overflow. But still, you will be able to lock out 99.99% of those script
    > kiddies who just try the overflow.
    > Maybe the percentage is not that accurate, but still the idea is
    > similar..
    
    I'd say from what I've just heard (16 responses in 12 hours - wow!) we can
    deduce the following:
    
    * non-root chrooted jails will stop 99.x% of buffer overflows due to the
      fact that the majority of such attacks are generic - and therefore rely on
      the presence of programs on the compromised systems to do their work.
      
    * there's a fair chance that a successful attack would need to be
      hand-crafted to work against your particular system. Congratulations,
      hacker has left the field for easier pickings ;-)
      
    I guess once in-memory (compared with executing local binaries) code
    execution becomes commonplace, this "advantage" will fade away.
    
    Still, nothing beats secure code to begin with.
    
    -- 
    Cheers
    
    Jason Haar
    
    Information Security Manager
    Trimble Navigation Ltd.
    Phone: +64 3 9635 377 Fax: +64 3 9635 417
    



    This archive was generated by hypermail 2b30 : Thu May 23 2002 - 21:11:57 PDT