In the cloud era, the DMZ has become more important – and more vulnerable – than its original architects ever thought possible. Ten or twenty years ago, back when most endpoints were still on-premise, the DMZ was an afterthought. It was very rare for a user outside the LAN to request access to a service within it, or vice versa, so the DMZ barely had anything in it. Now? Not so much.
Today, you’re either a software company, or you work with a huge number of SaaS vendors. You’re constantly providing access to users outside LAN or requesting access from services in the cloud. As a result, your DMZ is crowded with various applications. Although the DMZ is supposed to serve as a perimeter checkpoint, its function nowadays is more like an advertising billboard for attackers.
Each service you publish to the DMZ is another blip of information telling a potential hacker how many users you have, where you’re keeping your mission-critical data, and if that data includes something an attacker might want to steal. Here are four ways to stop that from happening.
- Make the DMZ a True Discontinuity
The idea behind the DMZ is that it must truly be separate from the LAN. As such, you should establish different IP routing and security policies in the DMZ, as opposed to the rest of the network. This makes it that much harder for attackers; they might figure out your DMZ, but they can’t then apply that knowledge to attack your LAN.
- Optimize Data Flows
Ideally, services from outside the DMZ will establish direct connections only to the DMZ itself. Services inside the DMZ will connect to the outside world only via proxies. Services inside the DMZ are more secure than those outside of it. Services that are better protected should assume the client role when requesting data from less-protected areas.
- Use a Two-Firewall Approach
While it’s possible to create a DMZ using just a single firewall with three or more network interfaces, two firewalls create a more secure deterrent. The first firewall represents the outer perimeter, and it directs traffic to the DMZ alone. The internal firewall allows traffic from the DMZ into the internal network. This approach is considered more secure, as it provides two distinct obstacles for an attacker to overcome.
- Implement Reverse Access
A patented approach from Safe-T, Reverse Access technology will make your DMZ even safer. This dual-server technology removes the need to open any ports within a firewall, while allowing secure application access between networks (through the firewall).
- External server – installed in the DMZ/external/non-secured segment.
- Internal server – installed in the internal/secured segment.
Located in the organization’s DMZ (on-premise or cloud), the role of the external server is to act as a front-end to all services/applications published to the Internet. It operates without the need to open any ports within the internal firewall and ensures that only legitimate session data can pass through into the internal network. The external server performs TCP offloading, allowing it to support any TCP based application without the need to perform SSL decryption.
The role of the internal server is to pull the session data into the internal network from the external SDA node, and only if the session is legitimate, perform layer 7 proxy functionality (SSL offloading, URL rewrite, Deep Packet Inspection, etc) and pass it to the destination application server.
Reverse Access technology works by allowing to authenticate accessing users before they have access to your mission-critical applications. An attacker with access to your applications over an illegitimate session can perform reconnaissance on your network, attempt, code injection attacks, or even try to move laterally through the network. Without the ability to pose as a legitimate session, however, an attacker’s toolkit becomes much more limited.
A poorly-implemented DMZ draws attackers like moths to a flame, but a DMZ with Reverse Access enabled is more like a bug zapper. Want to learn more about how to make this technology yours? Contact Safe-T today!
*** This is a Security Bloggers Network syndicated blog from Safe-T Blog authored by Julie Shafiki. Read the original post at: https://blog.safe-t.com/4-dmz-best-practices-to-shield-you-from-attackers