New firewalls should increase protection

This week, my company began deploying new firewalls. The old ones have been in place for more than six years; the new ones will allow us to take advantage of the next generation of features.

Today, application-based (Layer 7) firewalls provide far more flexibility than was available before. The methods of inspecting traffic enable us to allow or deny traffic based on a variety of factors. In addition, the firewall we chose, which is from Palo Alto Networks, offers what has been termed unified threat management (UTM), so we can eliminate several extra appliances and management consoles. UTM-type devices are not new, and in the past I'd found that all the functionality they offered had a big impact on performance. That's still a problem to some degree, but Palo Alto's system uses several chip sets for offloading and parallel-processing some of the functionality. That seems to minimize the performance hit to a satisfactory degree.

Trouble Ticket

At issue: Firewalls have gotten a lot more sophisticated since the company's were installed six years ago.

Action plan: Try out a new firewall system at some smaller sites, then embark on a companywide deployment if all goes well.

One thing about the new firewall technology that I like a lot is that it can be integrated with Active Directory, allowing us to build application-specific firewall rules based on individual needs. For example, if our remote-access policy didn't authorize the use of pcAnywhere but someone had a legitimate business need for it, I could write a rule and enable the use of that software, even by a single employee. At the same time, I could restrict that sort of remote access based upon time of day.

Naturally, the firewall offers URL content filtering for restricting access to certain Web sites and Web-based applications. But now we can do more than just block sites that traffic in porn, crime, terrorism and gambling; we can also define the sorts of activities that are permissible on some allowed sites. As things stand now, we give our employees access to third-party chat applications such as Yahoo Messenger, Google Talk and even Skype. But the firewall lets us prevent file transfers over such systems. That gives us a new way to further protect our intellectual property by stopping the illicit dissemination of sensitive documents. And again, if someone has a business need, I can make an exception.

Meanwhile -- and for me this just adds to the excitement -- we are directing the firewall logs to our new security event management tool. I hope to combine the rules from the firewall with rules from other application and server logs, as well as NetFlow traffic from our Cisco infrastructure, to provide meaningful information related to potential incidents. For example, if the firewall blocks a Port 80 connection to a malicious command-and-control server, we can correlate that with other data to determine how the user introduced malware, the origin of the infection and how many other resources were affected.

Not All at Once

For some of our smaller offices, we will also be enabling the intrusion-prevention feature (although we won't block traffic until we're confident that the rules are properly tuned). Other cool features include real-time detection and prevention of viruses and malware, traditional VPN options, and quality-of-service rules that will allow us to prioritize various types of traffic.

The initial rollout involves a half-dozen firewalls at some of our smaller sites. If we're happy with our experience, we'll continue the deployment in stages to the remaining 40-plus Internet points of presence that the company currently supports.

I am excited about this new technology and am hopeful for its success in our environment. But just in case, we're not getting rid of the old firewalls just yet.

This week's journal is written by a real security manager, "Mathias Thurman," whose name and employer have been disguised for obvious reasons. Contact him at mathias_thurman@yahoo.com.

Join in the discussions about security! computerworld.com/blogs/security