Best of Breed or Best Suite of Products

Should organizations implement layered defenses from different vendors? Should we rely upon a single vendor for an organization’s overall security?

According to a Gartner research paper, “Two firewall platforms are not better than one. We believe there is a higher risk associated with configuring and managing firewalls from multiple vendors than from a single vendor. Therefore, Gartner advises enterprises that have more than one firewall to standardize on a single vendor platform when the opportunity presents itself (that is, new installations or replacement during a refresh). In choosing a standard firewall, enterprises should consider the experience of their firewall administrators with each platform, scalability, central management and cost. ” (Young & Pescatore, 2008)

It also says that a firewall misconfiguration causes more than 99% of firewall breaches; not firewall flaws. It is true that debugging an error in any new appliance or tool can be cumbersome and time-consuming. Moreover, narrowing down to a single vendor relationship could help with greater discounts with less administration overhead.

However, there are situations where an enterprise could be stuck with a solution for long without much help to upgrade unless the enterprise pays almost the cost of a new solution and the extra cost of migrating to it. Sometimes it is better to diversify, especially when the industry is drastically changing and not all vendors address all issues with the changes. Continue reading “Best of Breed or Best Suite of Products”

Hardware or Host Based Firewalls

Do organizations need hardware firewalls when the network already has host-based software firewalls? Wouldn’t it add cost and complexity to networks? Wouldnt system protected by host-based software firewalls just as secure as having a hardware firewall if they are implemented appropriately?

“Firewalls actually come in two distinct flavors: software applications that run in the background and hardware devices that plug in between your modem and one or more PCs. Both types hide your PC’s presence from other systems, prevent unauthorized access from external sources, and keep tabs on network traffic across the firewall.” (Desmond, 2004)

The host-based software firewalls are good for the host; but not for the network that the host is connected to. A hardware-based firewall is required for:

  1. Network address translation (NAT) to prevent exposure of internal IP addresses,
  2. Port management to close unsolicited access to your host,
  3. Stateful packet inspection (SPI) to inspect for unsolicited incoming traffic,
  4. Virtual private network to support connection remote connection and the host,
  5. Activity logging and alerts
  6. Content and URL filtering

The hardware-based firewall is easy to implement and saves computing resources on the host. Malware on the host can bring down the firewall on the host, but not the hardware firewall.

While the hardware-based firewall can protect threats from outside the network, the software-based firewall helps to protect from attacks within the system. Software-based firewalls help to detect unauthorized outbound traffic from the host. A user can pick and choose which application can talk to peer hosts as well as external systems and may not be able to do this with hardware-based firewalls. Continue reading “Hardware or Host Based Firewalls”

Acknowledging Non-Applicable Threats

Is it important to account for or acknowledge risks that may not apply to an organization or system? What if you identified a risk that you would typically consider for but would not use due because of the context. Say, for example, your organization is not in a floodplain however it is usual to consider for the flood risk for all locations of your organization. What if you have validated with FEMA 100 Year Flood Zones that the total risk facing the organization is very low since it is not in a location that requires flood insurance? Do you still need to acknowledge the possibility of the threat occurring?

I believe it is essential to acknowledge the risk. We need to document it as very low risk, and very minimum safeguards are required as part of risk assessment. The building code of the location would define minimum safeguard. However, there could be situations where the asset value of the site is very high that you cannot ignore the risk altogether. Say the location is the primary data center for the organization. In such situations, the organization must implement all appropriate controls required to protect from the flood. The assessment needs to be revisited periodically to determine if the risk is significant or not at that time. Each evaluation must be based on current facts and numbers at the time. Continue reading “Acknowledging Non-Applicable Threats”