CVE
CVE-2023-31036

8.8HIGH

Key Information:

Vendor
nvidia
Status
Triton Inference Server
Vendor
CVE Published:
12 January 2024

Summary

A vulnerability exists in NVIDIA Triton Inference Server for both Linux and Windows platforms, triggered when the server is started with the non-default command line parameter --model-control explicit. This condition enables an attacker to exploit the model load API, potentially allowing them to perform a relative path traversal. Exploitation of this issue may result in unauthorized code execution, denial of service, privilege escalation, and the possibility of unauthorized access to sensitive information or manipulation of data.

Affected Version(s)

Triton Inference Server All versions prior to 2.40

References

https://nvidia.custhelp.com/app/answers/detail/a_id/5509

CVSS V3.1

Score:
8.8
Severity:
HIGH
Confidentiality:
High
Integrity:
High
Availability:
High
Attack Vector:
Network
Attack Complexity:
Low
Privileges Required:
Low
User Interaction:
None
Scope:
Unchanged

Timeline

  • Vulnerability published

  • Vulnerability Reserved

.