Researchers Uncovers Malware Behind Viasat Attack

Share post:

Viasat officials have confirmed the report by SentinelOne researchers that the malware, known as “AcidRain,” was used in a cyberattack that crippled thousands of Viasat customers’ modems.

AcidRain, a modern malware attributed to Russian state threat actors was discovered under the name “ukrop” for one of AcidRain’s source binaries.

According to SentinelOne researchers Juan Andres Guerrero-Saade and Max van Amerongen, the malware is an executable file for MIPS, the hardware architecture for the modems used by Viasat customers.

The researchers also discovered similarities between AcidRain and a “dstr,” the name of a wiper module for VPNFilter.

On Wednesday, Viasat confirmed the researchers’ investigation. According to Viasat, the attackers gained unauthorized access to a trust management segment of the company’s KA SAT network.

The attackers gained access by exploiting a misconfigured VPN after which they expanded their reach to other segments.

“The attacker moved sideways through this trusted management network to a specific network segment used to manage and operate the network, and then used this network access to execute legitimate, targeted management commands on a large number of residential modems simultaneously,” Viasat explained.

For more information read the original story in ArsTechnica.

SUBSCRIBE NOW

Related articles

Sleeper Supply Chain Attack Activates After 6 Years

A coordinated supply chain attack has compromised between 500 and 1,000 e-commerce websites by exploiting vulnerabilities in 21...

Russian-Controlled Open Source Tool Raises Alarms Over U.S. Cybersecurity

A widely used open-source Go library, easyjson, used in healthcare, finance and even defence has come under scrutiny...

Signal Archiving Tool Used By Trump Admin Is Breached, Raising Alarms Over Messaging Security (EDITORIAL)

(EDITORIAL) A messaging tool used by Trump administration officials to archive encrypted Signal messages has been hacked —...

Anthropic Warns: AI “Virtual Employees” Could Pose Security Risks Within a Year

Anthropic, a leading artificial intelligence company, anticipates that AI-powered virtual employees could begin operating within corporate networks as...

Become a member

New, Relevant Tech Stories. Our article selection is done by industry professionals. Our writers summarize them to give you the key takeaways