Channel Insider content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.

Testing an intrusion prevention system is still as much an art as a science, although the science is slowly gaining ground.

eWEEK Labs’ testbed for Top Layer Networks Inc.’s Attack Mitigator IPS 5500 combined an artificial, lab-created Internet connection with traffic carried by our ISP.

To get repeatable, comparable results, we also ran attack tools such as the open-source Nessus (www.nessus.org) on network devices while exposing them to the outside world. Using predictable attack traffic significantly speeds up proof-of-concept testing.

Effective testing of any IPS requires thorough knowledge of how an organization’s network is configured, both logically and physically. It is as important to know the protocols, applications and services running on a network as it is to know the routers, switches, firewalls and servers that are connected to a network.

We invited Top Layer to send an engineer to our San Francisco Labs site to help us install and configure the two Attack Miti- gator devices we used in our tests. Because an IPS is meant to be used in a high-performance, mission-critical environment, it’s not unusual to have vendor engineers involved early in the testing process. Security managers should expect to have at least several days of on-site support to ensure that the IPS is configured correctly.

Whenever possible, security managers should evaluate IPS devices in the environment in which they will be used. The listen-only mode of the Attack Mitigator and all the IPS devices we have tested allow the systems to operate with little to no noticeable impact on normal network performance. We think it’s worthwhile to put the IPS in-line right away.

Click here to read more about IPSes.

After getting the IPS devices in place, it’s advisable to let them run in listen-only mode for at least one week. When eWEEK Labs first implemented the Attack Mitigator devices, we ran them using mostly default rule configurations. We spent most of our time at that point viewing log files to see what traffic was coming in to the network and what was getting blocked by the IPS.

Whether you run IPSes in front of or behind firewalls depends on many factors. It’s a good idea to work with the IPS vendor to determine the best configuration for your organization.

We decided to run the Attack Mitigator devices in front of our firewalls because we wanted the IPSes to perform the bulk of the security work in our network.

Check out eWEEK.com’s for the latest security news, reviews and analysis. And for insights on security coverage around the Web, take a look at eWEEK.com Security Center Editor Larry Seltzer’s Weblog.