Kill Chain & the Internet of Things

The concept of a kill chain attack has been around for several years. The term originated from the military, but computer scientists at Lockheed-Martin Corporation were the first to use this term in the field of cybersecurity, describing a kill chain framework to defend computer networks in 2011. Its relevance has taken on new meaning in our current era of IoT devices and botnet attacks. IDC predicts that by 2020, 30 billion connected “things” will be a part of the digital infrastructure.

The  “kill chain” lays out the stages of a cyber-attack, starting from early reconnaissance to completion of the attack with the goal of data theft and enabling more attacks. These stages are:

1. Reconnaissance – The intruder selects its target device, researches it, and searches for vulnerabilities

2. Weaponization - Intruder uses a remote access malware weapon, such as a virus or worm, addressing a vulnerability  

3. Delivery - Intruder transmits weapon to the target device, whether through e-mail attachments, websites, USB drives, etc.

4. Exploitation - Malware weapons program code to triggers the attack. This then takes action on target network to exploit vulnerability.

5. Installation - Malware weapon installs access points for the intruder to use.

6. Command and Control – Malware then enables intruder to have "hands on the keyboard" persistent access to the target network, also enabling future attacks.

IoT devices, particularly items like security cameras, smart thermostats, wearables, and even coffee makers, are easy targets for kill chain intruders. They often have little or no security system, making step #2 of the kill chain rather easy.  For example, last year 80 Sony IP security camera models were found to have back doors, giving hackers easy access.

Don't Break the Kill Chain! Prevent it 
The best way to prevent a kill chain from infiltrating enterprise IoT security is to invest in a layered approach. There are four steps to this approach:

1. Assessment:  Start with a  network discovery process of all the existing IoT devices, including managed and partially managed devices. Understand what each type of device is, what operating system it is running on and which application and processes are installed on it.

2.  Segmentation:  IoT devices should not be in the same network segment as other devices, or within reach of the organization’s mission critical systems and data. Deploy firewalls between these segments to prevent “things”  from reaching the “crown jewels” of your network.

3. Detection:  Regularly analyze your network behavior to detect every IoT device which joins the network, and carefully examine if it behaves similarly to other typical devices. A compromised device or a fake device might look the same but behave differently.

4. Response:  Because manual alerts can take hours or even days to process, the best practice should involve some type of backup plan that will block or limit the access of a specific device within seconds.

This layered approach is designed to both prevent the likelihood of a kill chain attack, and also to break a live attack if one does occur. Once a vulnerability in the IoT device is detected and an attack is underway, breaking the final steps of the kill chain is most crucial, as it is often where the biggest gap lies in an organization’s advanced threat protection strategy. These last stages provide the best picture of who might be attacking and infecting your corporate network. They also require the least amount of time to remediate. For example, if a vulnerable security camera continues to communicate to an Internet forum, even after segmentation, it’s an easy call to block it entirely from the network. 

[Check out the two-day Dark Reading Cybersecurity Crash Course at Interop ITX, May 15 & 16, where Dark Reading editors and some of the industry's top cybersecurity experts will share the latest data security trends and best practices.]

Related Content: