VMware vCenter Server < 7.0 — RCE
An unauthenticated file upload vulnerability in the vCenter Server Analytics service allows arbitrary code execution. Apply VMware's patch immediately — automated exploitation tools were available within 24 hours of disclosure.
Unauthenticated root code execution on the vCenter Server, giving complete control of all managed ESXi hosts and VMs. Multiple public exploit tools were released within a day of the September 2021 disclosure, lowering the technical bar to mass exploitation.
The vCenter Server Analytics service accepts file uploads without authentication. An attacker with network access to port 443 on vCenter can upload a specially crafted file to the analytics service, which then executes the file on the server. The service runs as root on the vCenter appliance.
📧
Phishing link
🖼
Malicious file
🔓
Server compromised
Probably yes if any of these apply:
Affected OS versions
VMware rated this vulnerability 9.8 CRITICAL and issued an emergency advisory VMSA-2021-0020 on September 21, 2021. By September 22, automated scanning and exploitation had begun. Ransomware actors and cryptomining groups both exploited this vulnerability. The US CISA issued an urgent alert directing immediate patching. An exploit PoC was publicly available within 24 hours.
Manual remediation steps
⏱ 2 hours including appliance restartCheck Version and Exposure
# From vCenter Appliance Shell:
vmware -v
# Check if port 443 is reachable from internet (high risk):
# From an external machine:
curl -k -s "https://<vcenter-ip>/ui/" | head -5
Immediate Workaround (if patching delayed)
# Restrict access to vCenter port 443 at the firewall to management IPs only
# On vCenter appliance:
vc-support-bundle --file /etc/vmware/vsphere-ui/analytics.log
Apply the Patch
software-packages stage --iso /data/<patchfile>.iso --acceptEulas
software-packages install --staged
Verify
vmware -v
# Build must be 18455184 (7.0U2d) or 18831058 (6.7U3o) or later
No tested PowerShell script for this entry yet. We’re prioritising automation based on user demand.
References