IDS mailing list archives

Re: SourceFire RNA


From: Martin Roesch <roesch () sourcefire com>
Date: Wed, 3 Dec 2003 11:35:07 -0500

On Dec 2, 2003, at 12:17 PM, Lior Tal wrote:

Marty,
Many thanks for the reply.
When a computer is installed it usually includes many services that are
inactive and therefore passive detection may identify the device (IP and
OS) but it would be difficult or impossible to detect inactive services
that reflect open ports. These inactive services as far as I understand
still present vulnerabilities within the network.

They may, they may not. I don't know of any current vulnerabilities in echo or daytime, but MS RPC is another story.

Also, if you try to
mitigate the false alarms problem of NIDS sensors, is it possible to
tell whether an attack is going to be successful if you do not know of
these services.

Yes because if there are services available that are being attacked, we will detect their presence in real-time and deliver that information to our backend where the correlator is located, allowing us to correlate the information regardless of the order of arrival.

Another issue I have is the VA aspect of RNA presented on SourceFire's
web site - if passive detection can not detect all devices and running
services, how is it possible to provide reliable network map and
vulnerability information?

The same can be said of active discovery techniques, it is just as possible to hide from an active scanner as it is to hide from a passive one, so we can never know that we have 100% perfect knowledge of what's on our networks with either technology. On the other hand, I'm an advocate of the "perfect is the enemy of good enough" school of engineering, we need solutions that can detect changes in the network environment in real-time and scanners can't do that, RNA can and so it provides a good solution to a hard problem. Using that information to reduce the number of things that are hard to know about your network has value, and that's what we're building.

     -Marty


Kind regards,
Lior Tal

-----Original Message-----
From: Martin Roesch [mailto:roesch () sourcefire com]
Sent: Tuesday, December 02, 2003 6:27 PM
To: Lior Tal
Cc: focus-ids () securityfocus com
Subject: Re: SourceFire RNA

We can track and profile every active network element that's generating

traffic on the network and we can discover new elements in real-time.
The answer to the "how do you detect inactive hosts" question is "we
don't", you have to decide how important it is to know about machines
that are completely inactive on a network.  This kind of falls into the

"if a tree falls in the woods..." category from a certain standpoint,
but if you want to discover all the inactive hosts on your network and
track them on an ongoing basis then you can simply run an initial
discovery scan with any scanning tool (eg. nmap/strobe/hping/etc) and
RNA will see the scan traffic and auto-populate itself with host
representations for everything that responds.

      -Marty

On Dec 2, 2003, at 5:58 AM, Lior Tal wrote:



Hi,
Did anyone had a chance to evaluate the RNA published on SourceFire
web site?
From what I coule understand, they claim that by passive traffic
analysis the RNA can trace every network device, service and open port

within a network. It is difficult for me to understand how can passive

traffic analysis detect inactive devices and services which do not
transmit any network traffic?
Can anyone help figure that one?
Lior
US-Path Inc.


-----------------------------------------------------------------------
----

-----------------------------------------------------------------------
----


--
Martin Roesch - Founder/CTO, Sourcefire Inc. - (410)290-1616
Sourcefire: Intelligent Security Monitoring
roesch () sourcefire com - http://www.sourcefire.com
Snort: Open Source Network IDS - http://www.snort.org




--
Martin Roesch - Founder/CTO, Sourcefire Inc. - (410)290-1616
Sourcefire: Intelligent Security Monitoring
roesch () sourcefire com - http://www.sourcefire.com
Snort: Open Source Network IDS - http://www.snort.org


---------------------------------------------------------------------------
---------------------------------------------------------------------------


Current thread: