Re: Stateful Anomaly Detection Molding

• Previous message: Don Draper: "Re: Fortinet IDS"
• Maybe in reply to: Beauford, Jason: "Stateful Anomaly Detection Molding"
• Next in thread: Thomas Ptacek: "Re: Stateful Anomaly Detection Molding"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]

To: "Beauford, Jason" <jbeauford@EightInOnePet.com>, focus-ids@securityfocus.com
Date: Thu, 14 Oct 2004 08:38:03 -0500

>
> I want to get the groups opinion on the viability of Stateful Anomaly
> Detection Molding.
>

I've not heard the term "Stateful Anomaly Detection" before, but I presume
you are speaking of what I commonly call network profiling, or, more
precisely, behavior based anomaly detection. That said...

> With regards to IDS/IDP products which use S.A.D. as a detection engine,
> how easy/difficult is it to train the engine to allow malicious traffic.
> The idea is that these detection engines monitor traffic over a period
> of time and develop rule sets based on the given flow of traffic.

This question seems founded on the notion that there is One Way to do this
sort of monitoring, and I don't think that is a true premise. Most of the
research and actual implementations I've seen simply model a static view
of the network and compare with that over time. This could not easily be
"trained" to accept naughty behavior. Some do use a moving baseline of the
network, and this could indeed, at least conceptually, be trained. The
rapidity and effectiveness of the training would likely, however, be based
on the statistical methods used to construct that baseline, and this is
not likely to be by one common means.

> What can the Blackhats of the world due to perpetuate rule set molding
> of Stateful Anomaly Detection engines to allow malicious traffic through
> without being detected? How reliable are S.A.D. engines in detecting
> unwanted traffic?

Two big questions. One would presume that an effective profiling system
would detect even the smallest deviations from the norm, and an alert
analyst would take action. So, unless there was some deviation allowance
threshold, the attacker would have little room to maneuver. Thats really
the whole point of profiling, isn't it? As to the reliability, this is
too product specific to discuss generally, I suspect.

-ds

--------------------------------------------------------------------------
Test Your IDS

Is your IDS deployed correctly?
Find out quickly and easily by testing it with real-world attacks from CORE IMPACT.
Go to http://www.securityfocus.com/sponsor/CoreSecurity_focus-ids_040708 to learn more.
--------------------------------------------------------------------------

• Previous message: Don Draper: "Re: Fortinet IDS"
• Maybe in reply to: Beauford, Jason: "Stateful Anomaly Detection Molding"
• Next in thread: Thomas Ptacek: "Re: Stateful Anomaly Detection Molding"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]