Re: Security Audit

From: H C (keydet89at_private)
Date: Fri Sep 14 2001 - 19:03:38 PDT

  • Next message: Robert van der Meulen: "Re: 802.11B and libpcap"

    John,
    
    I appreciate your addition to the discussion. 
    However, I must say that I disagree with the idea of a
    "blind" anything, for several reasons.
    
    and foremost, it isn't safe.  What I mean by that is
    that if you're conducting a "blind" external pen test,
    and you have no idea what you're dealing with, with
    regards to the overall infrastructure, you could very
    well take down a mission-critical system.  
    
    Second, conducting a "blind" external pen test and
    telling the client that "I got in through this hole in
    that server" and telling him what patch to apply is
    doing him a disservice.  Doing so doesn't take a
    complete look at the overall infrastructure...you're
    only addressing one hole on one server.  The overall
    security of the entire infrastructure depends on a lot
    more than the sum of all the holes in all of the
    systems.  For example, when I look at an
    infrastructure with many NT systems, I'll take a look
    at the patch levels on each one.  Then, I'll analyze
    that information in the context of the entire
    infrastructure...what does it mean if the patch levels
    are all different?  How about if the patch levels are
    all the same, but they're all SP 4?  Doing so
    addresses the REAL security issues of the
    infrastructure.
    
    Third, a "blind" internal pen test provides as little
    meaningful information to the customer as an external
    one.  If you provide a list of holes on a list of
    systems, and what to do for each one (a la an ISS
    Internet Scanner report), you do nothing for your
    client that he couldn't do for himself...therefore,
    you add no value.  What are the real issues of the
    infrastructure?  A lack of staffing?  Is training
    needed?  Is it a lack of guidance or leadership from
    management.
    
    My point is simply this...customers pay consultants to
    provide a service.  That service should provide value,
    as well.  Anyone can purchase a commercial scanning
    product, and amortize the overall cost of the product
    and licensing over several clients.  The business
    differentiator for consulting firms is the analysis
    they provide.  In order to provide an analysis that is
    meaningful to and adds values to the customer, the
    consulting firm must understand the infrastructure as
    completely as possible.  This not only includes the
    technical aspects, but the day-to-day business
    processes, as well.  Pen tests do not provide this
    information.  
    
    Further, pen tests attempt to emulate a 'real world'
    attack, to some degree.  The attacker or pen tester
    will generally compromise a system with the first
    vulnerability that they successfully exploit.  If the
    pen tester finds a hole to get in, does he then go
    back and find all of the other possible holes?  Not
    likely.  So telling the customer how he got in and how
    to patch that hole does the customer little good. 
    They patch the hole and think they're safe.  Even if
    the consulting firm does a more comprehensive scan,
    and provides all of the holes they found that could be
    exploited, this is only a snapshot.  Instructing the
    customer to patch the holes adds no value.  A
    vulnerability assessment requires the consultant to
    understand the business processes, so that the
    recommendation can be provided in terms that describe
    security as a process.
    
    > Since we are all professionals, we 
    > usually gain access from the outside.  
    
    Of all of the statements you made, this one concerned
    me the most.  I'm not sure how being a professional
    equates to gaining access. 
     
    > When we walk into the CIO's office 
    > and hand him the administrative passwords we
    > instantly gain credibility 
    > while making the threat more tangible.  This
    > tangibility is part of the package. 
    
    What you're talking about here is really shock value. 
    You can obtain the same information in other
    ways...timing how long it takes for L0phtcrack to
    crack a certain percentage of the hashes in the SAM,
    for example.  
    
    Besides, what does it do to your credibility if you
    weren't able to gain access, let alone get admin
    passwords?   Do you, as a consultant, or your company
    offer a guarantee that you'll get in?  I'm just
    curious...
    
    Thanks,
    
    Carv
    
    
    
    
    
    
    __________________________________________________
    Terrorist Attacks on U.S. - How can you help?
    Donate cash, emergency relief information
    http://dailynews.yahoo.com/fc/US/Emergency_Information/
    
    ----------------------------------------------------------------------------
    This list is provided by the SecurityFocus Security Intelligence Alert (SIA)
    Service. For more information on SecurityFocus' SIA service which
    automatically alerts you to the latest security vulnerabilities please see:
    https://alerts.securityfocus.com/
    



    This archive was generated by hypermail 2b30 : Sun Sep 16 2001 - 23:07:37 PDT