Skip to main content

Botnets

Swarm of Bots

If I built a botnet, which I wouldn't, IT'S ILLEGAL, but I would use a Fast Flux network to hide my malware sites. My main targets would be IoT devices, and Wi-Fi routers.

With the end goal of a Botnet mining Bitcoin via the GPU / CPU resources of infected devices.

Fast Flux Networks​

Is a DNS technique used to hide phishing and malware delivery sites behind an ever-changing network of compromised hosts acting as proxies. It involves rapidly changing the DNS records (A or MX records) to mix up IP addresses frequently.

This networking layer makes it difficult for law enforcement and security professionals to locate the rogue servers.

P2P Architecture​

P2P, or (peer-to-peer) gets a lot of fanfare in the blockchain world. For Botnets it's how to build resiliency into a system where each agent acts as both a client and a server.

A P2P architecture can be orchestrated to decentralize a command and control infrastructure, allowing "nodes" to perform missions and if a device is caught and "cleaned" makes it harder to take down the entire botnet.

Specific nodes can be given more control or priority in the decision-making process, either because they are more trusted by the bot herder or they have more resources available. This can be particularly useful for distributing commands, updates, or propagating new malware to the whole botnet.

IOT Devices​

Many IoT devices are poorly secured. Why? The majority of older IOT devices have login credentials that were never changed. They run a stripped down linux kernel operating system. Why does this matter?

These devices can be easily compromised. An average of 22 "IoT" devices live inside the American home.

Wi-Fi Routers​

IoT devices connect to the internet via the home Wi-Fi router, or send a signal to a "bridge device" or even directly pair with your smart phone. If an IoT device is compromised depending on its privileges.

It can be used to attack or probe other devices on the same network, including the router. At the very least it can sniff traffic, and capture credentials, to gain access to the router.

Propagation Techniques​

Worms​

Self-propagating malware can spread to other systems automatically. If there's an exploitable vulnerability present on many systems, a worm can be an effective way to build a botnet quickly.

Trojans​

Malware disguised as legitimate software can be inadvertently installed by users, adding their machines to a botnet.

Drive-by Downloads​

Simply visiting a malicious website or viewing a malicious ad can lead to malware being silently downloaded and executed on a user's system.

Phishing​

The age old tricking users into executing malicious files or giving away credentials is still the most common way to compromise systems.

Botnet Objectives​

Once a botnet is established, it can be used for various malicious activities, including DDoS attacks, spamming, cryptocurrency mining, and more.

Defensive Measures​

Change Default Credentials​

Always replace default usernames and passwords on both IoT devices and routers.

Firmware Updates​

Regularly update the firmware on devices and routers to patch known vulnerabilities.

Network Segmentation​

If possible, separate IoT devices from primary devices using different network segments or VLANs.

Secure Configurations​

Disable unnecessary services, ensure encryption is used for communication, and employ strong, unique passwords.

Attack Vectors​

Wi-Fi router passwords, especially default ones, can be a significant vulnerability if not changed Let's break it down:

Brute-Forcing Wi-Fi Passwords​

WPA and WPA2 encryption for Wi-Fi have proven resistant to brute-forcing, especially if strong, long passwords are used. However, WEP, an older standard, can be cracked quickly.

The primary attack vector against WPA/WPA2 is the WPS PIN, if enabled. It's an 8-digit number, and due to its structure, it's effectively reduced to (10^4 + 10^3) possibilities, making brute-forcing feasible.

WPA/WPA2 can be cracked using the 4-way handshake and then using offline brute force or dictionary attacks. This method's effectiveness depends on the password's strength.

Conclusion​

Constructing and maintaining a botnet targeting IoT devices is a formidable undertaking that involves significant risks, many complexities, and illegal activities.

The primary purpose of using GPU/CPU resources of infected devices for cryptocurrency mining – faces practical challenges. This includes the limited processing power of most IoT devices and the inefficiency of CPU mining for popular cryptocurrencies.

However, considering the variance in Exploit Kits - I explore the idea further.