There really is a difference between monitoring a network
(focus is primarily on the src-dst pairings) and an IDS (which focuses on the
contents of packets and port usage). The more you ask a system to do
per-packet, the fewer packets can be analyzed. That's the trade off.
Another is placement - an IDS is usually either part of the firewall or a
passive observer of the traffic to/from the firewall. Where as a
monitoring system is more frequently placed INSIDE the
firewall.
However, there are a few stories in the back traffic of the
ntop lists on using ntop to react/identify problems. IIRC, it was at
the time of one of the email worms - the user brought up a second instance of
ntop with a "port 25" filter - that showed all the mail 'clients' and the top
one was (naturally) the infected host. Bada-bing,
bada-boom...
-----Burton
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Riccardo Rosso
Sent: Friday, April 15, 2005 9:12 AM
To: [email protected]
Subject: [Ntop] ntop as IDS
Thanks to both of you.
I really do know and use Snort since almost 5 years but I was required to evaluate ntop as a NIDS and apart from these as you know Luca Deri wrote a lot on this subject (and on the "burglar alarm" option of ntop) so I was also personally inetersted in the subject but unfortunately it seems that nobody (nor Stef who originally replied to me mentioning its fine study for GIAC) had used ntop before as a NIDS nor knew how to configure it.
Thank again,
hope some guy will reply to me (maybe Luca) on the subject sooner or later,
best regards,
Riccardo
<snip />
_______________________________________________ Ntop mailing list [email protected] http://listgateway.unipi.it/mailman/listinfo/ntop
