Skip to main content
a promotional graphic telling you that PCI Compliance is no longer an annual exercise and that continuous monitory must be built in
Locking Down vCenter and ESXi Before Ransomware Finds ThemCyber Threats & Attacks
4 min readFor Chief Information Security Officers (CISOs)

Locking Down vCenter and ESXi Before Ransomware Finds Them

The Problem: Why This Matters

Kaspersky's Global Emergency Response Team (GERT) recently revealed how a management-plane attack unfolds. Over ten days, a group called BlackNevas destroyed an organization's virtualization setup: four ESXi hosts, all VMs, both vCenter Server Appliances, and connected backup infrastructure. They didn't exploit any hypervisor vulnerabilities; instead, they used legitimate VMware management interfaces.

Two factors make this scenario likely to repeat. First, the leaked Babuk source code, including its ESXi version, allows even low-skill operators to deploy ESXi lockers. At least ten ransomware families have used it. Second, ransomware made up 39.4% of incidents GERT handled in 2025. Simply patching won't stop an attacker with valid credentials. This guide focuses on hardening those access paths.

What You Need Before Starting

  • Inventory: List every ESXi host, vCenter appliance, and backup server, noting ESXi version and patch level. Unsupported versions should be retired.
  • Admin Access: Ensure access to vCenter, ESXi hosts, AD, and backup platform.
  • Network Control: Manage VLANs and use a hardened jump host.
  • Log Destination: Use a SIEM or syslog collector outside the virtualization estate.
  • Change Window and Backup Target: Have an offline or immutable backup target ready for testing.
  • Named Owners: Assign owners for virtualization admin, identity, and backup roles.

Step-by-Step Implementation

1. Isolate the Management Plane. Place ESXi management interfaces, vCenter, and backup servers on a dedicated segment. Allow inbound access only from a hardened jump host. Block SSH and the vSphere Client from general workstations and server VLANs.

2. Break Credential Reuse. Create admin accounts specifically for virtualization. Domain controller, jump server, backup, and workstation admin credentials shouldn't access vCenter or ESXi. This separation is crucial, as the incident involved credentials obtained from non-virtualization systems.

3. Remove Direct AD Dependency from ESXi. Unjoin hosts from AD where possible, using vCenter RBAC with identity federation. If AD integration is necessary, address issues related to CVE-2024-37085. Broadcom's workaround for older systems includes setting esxAdminsGroupAutoAdd to false and adjusting authValidateInterval. Patching to ESXi 8.0 U3 fixes the vulnerability, but these settings add defense in depth.

4. Shrink the SSH Footprint. Disable ESXi SSH and Shell services by default, enabling them only for maintenance. Enable lockdown mode so hosts accept management only through vCenter. Restrict appliance shell access to named accounts and alert on every enablement.

5. Treat the Backup Server as a Tier-Zero Asset. Backup platforms hold critical credentials and network access. Avoid domain-joining the backup server. Rotate stored service credentials, require MFA for the console, and keep at least one backup copy immutable or offline.

6. Close Known Entry Points. Map your estate against vulnerabilities like CVE-2021-21972 and CVE-2021-22005. Prioritize unauthenticated remote code execution paths.

7. Fix the Visibility Gap. Forward ESXi host logs and vCenter logs to an external collector. Add network-side telemetry and agentless integrations. In the incident, host-native and management-plane telemetry were key for timeline reconstruction.

8. Build Detections for Valid-Account Abuse. Alert on:

  • vCenter or ESXi logins from unusual sources.
  • Logins outside known admin hours.
  • Changes to ESXi permissions or group mappings.
  • SSH or Shell service enablement.
  • Bulk VM power-off or datastore operations.

Validation: How to Verify It Works

  • Reachability Test: From a standard workstation and server VLAN, attempt connections to the vSphere Client, SSH, and the ESXi management interface. All should fail.
  • Credential Test: Try authenticating to vCenter with a domain admin account and a backup service account. Both should be denied.
  • AD Test: Confirm membership in the old ESX Admins group grants no host rights.
  • Detection Test: Enable SSH on a lab host and log in from an unexpected source. Confirm your SIEM raises an alert promptly.
  • Restore Test: Recover a representative VM, including a domain controller, from the immutable copy, using credentials that don't exist in production.
  • Tabletop: Walk your team through a 10-day dwell scenario and check whether your logs would reconstruct it.

Maintenance: Ongoing Tasks

  • Monthly: Review vCenter permissions and ESXi local accounts; reconcile against the admin roster.
  • Per Advisory: Patch or mitigate new vCenter and ESXi vulnerabilities, focusing on unauthenticated RCE.
  • Quarterly: Rotate virtualization and backup service credentials, and rerun reachability and credential tests.
  • Twice a Year: Run a full restore exercise from immutable backups.
  • Continuously: Retire unsupported ESXi versions.

For risk and insurance, underwriters increasingly examine segmentation, MFA, and backup isolation. The evidence from these tests supports accurate answers, protecting you from Application Fraud Warranty exposure at claim time.

Promotional banner for the Pentest Readiness checklist download

You Might Also Like