NVIDIA Triton Inference Server Denial of Service

Executive Summary

CVE-2026-47480 is a vulnerability in NVIDIA Triton Inference Server for Linux. It involves an uncaught exception, which means the software fails to handle an unexpected error condition properly.

An attacker can exploit this vulnerability by triggering the uncaught exception, which may cause the server to crash or become unresponsive. This results in a denial of service (DoS), where legitimate users are unable to access the service.

• The vulnerability is classified under CWE-248 (Uncaught Exception).
• It affects Triton Inference Server versions up to and including 26.04.
• The CVSS v3.1 base score is 7.5 (HIGH), with a vector of AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H, indicating it can be exploited remotely without authentication.

Useful 0 Not Useful 0

Detection Guidance

Detecting this vulnerability on your network or system requires checking if you are running a vulnerable version of NVIDIA Triton Inference Server for Linux. The vulnerability affects versions up to and including 26.04.

• Check the installed version of Triton Inference Server by running the following command: tritonserver --version or by inspecting the package manager, e.g., dpkg -l | grep triton or rpm -qa | grep triton.
• Monitor network traffic for unusual patterns or unexpected requests targeting the Triton Inference Server, particularly on the ports it uses for inference services (default is often 8000, 8001, or 8002). Tools like Wireshark or tcpdump can be used for this purpose.
• Review server logs for uncaught exceptions or crashes, which may indicate exploitation attempts. Look for entries in /var/log/ or the specific log directory configured for Triton Inference Server.

Useful 0 Not Useful 0

Impact Analysis

If you are using NVIDIA Triton Inference Server for Linux, this vulnerability could impact you in the following ways:

• Denial of Service (DoS): An attacker could crash the server, making it unavailable for legitimate users or applications that rely on it for inference tasks.
• Service Disruption: If the server is part of a critical workflow (e.g., real-time data processing, AI model serving), its unavailability could disrupt operations or cause downtime.
• No Data Theft or Modification: The vulnerability does not allow for unauthorized access to data (confidentiality) or modification of data (integrity), as indicated by the CVSS vector (C:N/I:N).

Useful 0 Not Useful 0

Compliance Impact

The impact of this vulnerability on compliance with standards and regulations depends on how the affected system is used:

• GDPR: If the Triton Inference Server processes personal data of EU citizens, a denial of service could lead to availability issues. GDPR requires ensuring the availability and resilience of processing systems (Article 32). Prolonged downtime due to this vulnerability could be seen as a failure to meet these requirements.
• HIPAA: For organizations handling protected health information (PHI), HIPAA requires ensuring the availability of systems that store or process PHI. A DoS attack exploiting this vulnerability could disrupt access to critical healthcare services, potentially violating HIPAA's availability requirements.
• Other Standards: Compliance frameworks like ISO 27001 or SOC 2 emphasize the need for system availability and resilience. A successful exploit of this vulnerability could indicate a failure to implement adequate controls for ensuring service continuity.

However, since this vulnerability does not involve data breaches or unauthorized access, the primary compliance risk is related to service availability rather than data confidentiality or integrity.

Useful 0 Not Useful 0

Mitigation Strategies

To mitigate this vulnerability, follow these immediate steps:

• Upgrade NVIDIA Triton Inference Server to a version beyond 26.04, if available, as the vulnerability affects versions up to and including 26.04. Check NVIDIA's official documentation or support channels for the latest patched version.
• If an upgrade is not immediately possible, restrict network access to the Triton Inference Server to trusted IP addresses only. Use firewalls or network security groups to limit exposure to potential attackers.
• Monitor the server for signs of exploitation, such as unexpected crashes or denial of service, and implement rate limiting or request filtering to reduce the risk of exploitation.
• Apply any available workarounds or patches provided by NVIDIA, even if a full upgrade is not feasible. Check NVIDIA's security advisories for additional guidance.

Useful 0 Not Useful 0