A new bot called DemonBot is targeting Hadoop clusters to execute distributed denial-of-service (DDoS) attacks.

The Radware Threat Research Center recently observed a threat actor exploiting a Hadoop Yet Another Resource Negotiator (YARN) unauthenticated remote command execution. This method of attack enables the malicious agent to infect clusters of Hadoop, an open source distributed processing framework that helps big data apps run in clustered systems, with DemonBot. Upon successful infection, the threat connects to its command-and-control (C&C) server and transmits information about the infected device.

Why Cloud Infrastructure Servers Are Juicy Targets

The threat’s goal is to leverage infected cloud infrastructure servers to conduct DDoS attacks. At this juncture, it is not exhibiting worm-like behavior akin to Mirai. Instead, it relies on 70 exploit servers for distribution, infrastructure that helps it perform 1 million exploits every day.

That being said, Radware found DemonBot to be binary-compatible with most Internet of Things (IoT) devices, which means the threat could spread to other types of products.

DemonBot isn’t the first bot to target cloud infrastructure servers like Hadoop clusters. In early October, a security researcher reported on Twitter that handlers of the Sora IoT botnet attempted to exploit the same YARN abused by DemonBot.

Radware attributed the growing interest in Hadoop to the fact that cloud infrastructure servers allow bad actors to stage larger and more stable DDoS attacks using multiple vectors, such as User Datagram Protocol (UDP) and Transmission Control Protocol (TCP) floods.

How to Defend Against DemonBot

Security professionals can help protect their organizations against DemonBot by conducting a proper risk assessment on their cloud deployment. From there, they should enlist the help of penetration testers to map the vulnerabilities affecting their deployment.

Security teams should also look to invest in mitigation tools and services that specialize in defending against a DDoS attack.

Sources: Radware, Ankit Anubhav