BlackNevas wiped an ESXi estate via VMware management tools
Original title: Ransomware at the hypervisor layer: how attackers use ESXi and vCenter features for malicious purposes
Over 10 days the attacker hit four ESXi hosts, every guest, both vCenters and connected backups, with no hypervisor exploit.
Kaspersky’s GERT investigated a BlackNevas ransomware engagement that destroyed an organization’s virtualization estate through legitimate VMware management interfaces. No guest-to-host escape or hypervisor memory corruption was used. The write-up links broader ESXi locker availability to the September 2021 Babuk source leak and says targeted organizations often run unsupported ESXi. Guest EDR cannot see datastore encryption performed by the hypervisor.
Unconfirmed: The BlackNevas name comes from ransom notes; the article does not state how this intrusion first reached the management plane.
- Ransom notes named BlackNevas, a group first observed in November 2024.
- Dwell time from initial access to destruction was 10 days.
- At least ten later families reused the leaked Babuk ESXi code.
- Kaspersky says ransomware was 39.4% of GERT incidents in 2025.
- Off-the-shelf agents are not supported on ESXi or the vCenter Server Appliance.
Impact: Virtual machines and backup infrastructure were destroyed
| Product | Affected |
|---|---|
| VMware ESXi | — |
| VMware vCenter Server Appliance | — |
- 1.Collect host-native and management-plane telemetry; guest EDR cannot attribute ESXi datastore changes.
- Products
- ESXivCenter Server Appliance
- Threat actors
- BlackNevas
- Malware
- Babuk
- Organizations
- Kaspersky
From the original article
12 min read at securelist.com
Kaspersky experts analyze a recent ransomware incident and discuss why the hypervisor layer remains difficult to monitor, what common access paths adversaries use to reach it, and how to protect your ESXi systems.
By Elsayed Elrefaei, Mohamed Ghobashy · securelist.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.