RE: Low cost HID based IDS system

From: Zach Forsyth (Zach.Forsyth_at_kiandra.com)
Date: 05/19/03

  • Next message: Andrew Plato: "Re: Network Taps"
    Date: Mon, 19 May 2003 10:21:01 +1000
    To: "Schmehl, Paul L" <pauls@utdallas.edu>, "Focus-Ids" <focus-ids@securityfocus.com>
    

    Paul,

    You seemed to of missed the point a little.
    Why do people bother developing snort when there are so many other
    commercial IDS's out there, it's free so therefore it can't be any good.
    Why do people bother with Nessus
    Why do people bother with <insert free/cheap/open source solutions here>

    Why would anyone try to build something lower cost that what is already
    out there? Silly developers/engineers.
    Come on, give me a break, I would like to explore the possibility of
    implementing a low cost solution for my clients using relatively
    inexpensive solutions.

    They realise that I can only offer them 8am-7pm full coverage during
    business hours, occasional after hours and weekend monitoring.
    They still get event analysis the next morning from any of the nights
    activities, and this is still valuable to them.
    Hopefully when we have enough clients to warrant the expense we can do
    24x7. And that would be never trying to implement a high cost solution
    as the clients wouldn't pay.

    What's better 8am-7pm monitoring and further analysis or nothing?

    I believe I can roll out a better than nothing solution to small
    clients, say of 5 pc's and a server for a small monthly fee.
    Lets try and work through some ideas, this are pretty crazy so bear with
    me:

    Let's assume they do not have US$1000 per month.

    Buy a nice fat NID and sit it at an internet gateway. Lets say we have a
    10mb connection to the internet. Lets say a Cisco 4210 for US$10,000
    (just adding a rough figure here)
    All clients then use you as an ISP for their internet/mail/etc.
    We now have a gateway IDS device protecting multiple clients for a
    shared cost to each client per month.
    How many small clients could we connect? I would put it between 50-X
    clients depending on cost.
    If we assumed all of them would of normally had an ISDN or modem
    connection to the internet then it could be very high. And there are
    still plenty of those clients around.
    Would they all pay $50/100/150,200 a month?
    We could make it profitable as long as they realise it is better than
    nothing but not as good as a full 24x7 service.
    But then again what are they paying per month...

    50 clients x$50 = $2500 per month
    50 clients x$100 = $5000 per month

    Seems to me like the NID will be payed off nice and quickly, then the
    cost is based on monitoring hours, reporting and incident response
    (which could be charged separately)

    What about a HID based solution?
    Buy a central event monitoring and management console. Lets say
    Enterasys Dragon US$7000 or so
    Deploy a server based HID agents to each clients critical server.
    Enterasys squire US$750 or so
    Then once again we can protect how many HID's? Some event monitoring
    consoles will support 10,000's of remote agents.
    We now have a managed IDS for small clients, that can pay a small fee
    per month for 7am-8pm protection.

    What about an IPS which still needs 24x7 monitoring in large corporate
    clients, but can still protect you 24x7 by itself if you don't have the
    monitoring.
    I think you could build a relatively cheap IPS solution with monitoring
    and reporting for small clients.
    This may even be the way I am headed as it offers more protection for
    the $ spent. Something like Cisco Secure Agent, or Entercept.

    As long as clients realise what they are paying for and receiving for
    their money I think a simple yet small solution will work.
    Value for money is relative and I think usually as price goes up you get
    to a point of diminishing returns.
    It is fine for large corporate clients to strive for that last 1% but
    small companies get no value out of doing that.

    I appreciate your scepticism and it has helped me think through some
    aspects of this in more detail.

    I guess it all depends on your viewpoint of what an IDS should be.
    I can see a big gap in the market and will continue to work on
    innovative ways to provide solutions for all the small fish out there.

    Cheers

    Zach

    -----Original Message-----
    From: Schmehl, Paul L [mailto:pauls@utdallas.edu]
    Sent: Saturday, 17 May 2003 14:10 PM
    To: Alan Shimel; Zach Forsyth; Focus-Ids
    Subject: RE: Low cost HID based IDS system

    Nothing in life is free. Everything has a cost associated with it. For
    example, while he may be able to provide a similar service for much less
    money, what happens if he misses an attack that devastates one of his
    customer's networks? Will he indemnify them? Will they sue him and
    destroy *his* business in the process? Is he going to be watching the
    IDS 24/7 like an MSSP would? Is he knowledgeable enough of IDS to
    provide the same level of service to them that an MSSP would? Does he
    have the resources?

    Everything has a cost. Sometimes the cost doesn't show up until you've
    already realized the decision you made was flawed. What's the value of
    the business lost while your network is down?

    I just don't think it makes good business sense to cut corners on
    security to save a few dollars. In the end, you'll regret it. ISTM he
    would serve his customers better by negotiating a reasonable rate for
    the services of an MSSP *through* his company to each of his customers.
    With his higher bargaining power, he has the opportunity to provide them
    with real value at a reasonable cost that is much less than what they
    might be able to negotiate on their own. Especially now, when security
    companies are scrambling to find revenue.

    In the final analysis the question he needs to answer is; is he trying
    to provide his customers with true value for their dollars? Or just
    throw together something cheap that will make them feel safer but won't
    really make them any more secure?

    Paul Schmehl (pauls@utdallas.edu)
    Adjunct Information Security Officer
    The University of Texas at Dallas
    AVIEN Founding Member
    http://www.utdallas.edu/~pauls/

    -----Original Message-----
    From: Alan Shimel [mailto:alan@latis.com]
    Sent: Friday, May 16, 2003 10:00 PM
    To: Schmehl, Paul L; Zach Forsyth; Focus-Ids
    Subject: RE: Low cost HID based IDS system

    There are tools out there that would allow him to provide these services
    to customers at substantially below some of the MSSPs you mentioned
    charged. I think it is possible to provide this service sub-1000
    dollars a month

    -------------------------------------------------------------------------------
    INTRUSION PREVENTION: READY FOR PRIME TIME?

    IntruShield now offers unprecedented Intrusion IntelligenceTM capabilities
    - including intrusion identification, relevancy, direction, impact and analysis
    - enabling a path to prevention.

    Download the latest white paper "Intrusion Prevention: Myths, Challenges, and Requirements" at:
    http://www.securityfocus.com/IntruVert-focus-ids2
    -------------------------------------------------------------------------------


  • Next message: Andrew Plato: "Re: Network Taps"

    Relevant Pages

    • Re: Eyes In The Sky
      ... Ian Erickson could probably fetch a photo of ... for little to no cost. ... You call in sick. ... I'm talking in these cases clients who had ...
      (misc.survivalism)
    • Re: DST
      ... With the cost of college for the 2.3 kids, vehicles, ... car payments, insurance, groceries, clothes, etc. for two people, all ... and most of his clients are "vintage". ...
      (rec.woodworking)
    • Re: Electricity!
      ... the cost of migrating to a new machine is REALLY HIGH. ... dynamic data that was downloaded (mostly tax forms and tax tables, ... And it cost about $800 in my time to do the migration. ... program reasonably for my clients. ...
      (soc.motss)
    • Re: Building Contract. Please help.
      ... The clients provided a drawing which was ... increases that they would increase the final cost by £X? ... roof without scaffolding either. ... Post a copy of your contract then you will get a realistic answer ...
      (uk.legal)
    • Re: Migrating SBS servers
      ... internet into the charges against my clients. ... cost is shared between all clients. ... customer, you leave everything behind and take nothing with you. ... Therefore many of us do recommend a swing migration ...
      (microsoft.public.windows.server.sbs)