NVIDIA DCGM Exporter Bug Exposes GPU Servers to Resource Exhaustion Attacks

What happened?

NVIDIA patched high-severity flaw CVE-2026-47483 in DCGM Exporter that allows unauthenticated remote attackers to exhaust server resources, crash GPU monitoring, and disrupt AI workloads. DCGM Exporter is a common monitoring tool for GPU servers and Kubernetes clusters. It reads details such as GPU model, temperature, memory use, power draw, errors, and workload activity, then serves the data in a Prometheus-friendly format over HTTP. A unique UUID also lists each GPU.

Key details

• NVIDIA fixed a high-severity flaw in DCGM Exporter that lets unauthenticated remote attackers exhaust server resources and crash GPU monitoring.
• Attackers can send many profiling requests at once to cause uncontrolled memory and CPU use until the exporter slows, stops responding, or crashes.
• NVIDIA rated the issue 8.2 out of 10 and classified it as CWE-770 for resource allocation without proper limits.
• Shodan scans found over 2,100 exposed DCGM Exporter hosts revealing more than 12,000 GPU UUIDs across about 300 organizations.
• Exposed hardware includes H100, H200, B200, Blackwell Ultra B300, RTX 4090, and RTX 5090 GPUs, with the United States accounting for 44 percent of observed GPUs.
• Customer-run endpoints were found on networks linked to GPU cloud providers such as Voltage Park, Lambda, Northern Data, and DigitalOcean.

Summarised by SecOraLens from Cyber Security News. Always read the original source before acting. How SecOraLens verifies advisories.