Vulnerability-Lookup

CNVD-2025-18930

Vulnerability from cnvd - Published: 2025-08-20

Title

NVIDIA Triton Inference Server Python后端越界写漏洞

Description

Triton Inference Server是NVIDIA开发的一款高性能推理服务引擎，专为生产环境中的AI模型部署设计，支持多种框架（TensorFlow、PyTorch、ONNX 等），并优化了GPU和CPU的推理性能。 NVIDIA Triton Inference Server Python后端存在越界写漏洞，该漏洞源于内存缓冲区数据越界被写，远程攻击者可利用该漏洞提交特殊的HTTP请求，可以应用程序上下文执行任意代码。

Severity

高

Patch Name

NVIDIA Triton Inference Server Python后端越界写漏洞的补丁

Patch Description

Formal description

目前厂商已经发布了升级补丁以修复这个安全问题，请到厂商的主页下载: https://github.com/triton-inference-server/server/releases

Reference

https://nvidia.custhelp.com/app/answers/detail/a_id/5687

Impacted products

Name
NVIDIA Triton Inference Server <25.07

Show details on source website

JSON

To clipboard

{
  "cves": {
    "cve": {
      "cveNumber": "CVE-2025-23319",
      "cveUrl": "https://nvd.nist.gov/vuln/detail/CVE-2025-23319"
    }
  },
  "description": "Triton Inference Server\u662fNVIDIA\u5f00\u53d1\u7684\u4e00\u6b3e\u9ad8\u6027\u80fd\u63a8\u7406\u670d\u52a1\u5f15\u64ce\uff0c\u4e13\u4e3a\u751f\u4ea7\u73af\u5883\u4e2d\u7684AI\u6a21\u578b\u90e8\u7f72\u8bbe\u8ba1\uff0c\u652f\u6301\u591a\u79cd\u6846\u67b6\uff08TensorFlow\u3001PyTorch\u3001ONNX \u7b49\uff09\uff0c\u5e76\u4f18\u5316\u4e86GPU\u548cCPU\u7684\u63a8\u7406\u6027\u80fd\u3002\n\nNVIDIA Triton Inference Server Python\u540e\u7aef\u5b58\u5728\u8d8a\u754c\u5199\u6f0f\u6d1e\uff0c\u8be5\u6f0f\u6d1e\u6e90\u4e8e\u5185\u5b58\u7f13\u51b2\u533a\u6570\u636e\u8d8a\u754c\u88ab\u5199\uff0c\u8fdc\u7a0b\u653b\u51fb\u8005\u53ef\u5229\u7528\u8be5\u6f0f\u6d1e\u63d0\u4ea4\u7279\u6b8a\u7684HTTP\u8bf7\u6c42\uff0c\u53ef\u4ee5\u5e94\u7528\u7a0b\u5e8f\u4e0a\u4e0b\u6587\u6267\u884c\u4efb\u610f\u4ee3\u7801\u3002",
  "formalWay": "\u76ee\u524d\u5382\u5546\u5df2\u7ecf\u53d1\u5e03\u4e86\u5347\u7ea7\u8865\u4e01\u4ee5\u4fee\u590d\u8fd9\u4e2a\u5b89\u5168\u95ee\u9898\uff0c\u8bf7\u5230\u5382\u5546\u7684\u4e3b\u9875\u4e0b\u8f7d:\r\nhttps://github.com/triton-inference-server/server/releases",
  "isEvent": "\u901a\u7528\u8f6f\u786c\u4ef6\u6f0f\u6d1e",
  "number": "CNVD-2025-18930",
  "openTime": "2025-08-20",
  "patchDescription": "Triton Inference Server\u662fNVIDIA\u5f00\u53d1\u7684\u4e00\u6b3e\u9ad8\u6027\u80fd\u63a8\u7406\u670d\u52a1\u5f15\u64ce\uff0c\u4e13\u4e3a\u751f\u4ea7\u73af\u5883\u4e2d\u7684AI\u6a21\u578b\u90e8\u7f72\u8bbe\u8ba1\uff0c\u652f\u6301\u591a\u79cd\u6846\u67b6\uff08TensorFlow\u3001PyTorch\u3001ONNX \u7b49\uff09\uff0c\u5e76\u4f18\u5316\u4e86GPU\u548cCPU\u7684\u63a8\u7406\u6027\u80fd\u3002\r\n\r\nNVIDIA Triton Inference Server Python\u540e\u7aef\u5b58\u5728\u8d8a\u754c\u5199\u6f0f\u6d1e\uff0c\u8be5\u6f0f\u6d1e\u6e90\u4e8e\u5185\u5b58\u7f13\u51b2\u533a\u6570\u636e\u8d8a\u754c\u88ab\u5199\uff0c\u8fdc\u7a0b\u653b\u51fb\u8005\u53ef\u5229\u7528\u8be5\u6f0f\u6d1e\u63d0\u4ea4\u7279\u6b8a\u7684HTTP\u8bf7\u6c42\uff0c\u53ef\u4ee5\u5e94\u7528\u7a0b\u5e8f\u4e0a\u4e0b\u6587\u6267\u884c\u4efb\u610f\u4ee3\u7801\u3002\u76ee\u524d\uff0c\u4f9b\u5e94\u5546\u53d1\u5e03\u4e86\u5b89\u5168\u516c\u544a\u53ca\u76f8\u5173\u8865\u4e01\u4fe1\u606f\uff0c\u4fee\u590d\u4e86\u6b64\u6f0f\u6d1e\u3002",
  "patchName": "NVIDIA Triton Inference Server Python\u540e\u7aef\u8d8a\u754c\u5199\u6f0f\u6d1e\u7684\u8865\u4e01",
  "products": {
    "product": "NVIDIA Triton Inference Server \u003c25.07"
  },
  "referenceLink": "https://nvidia.custhelp.com/app/answers/detail/a_id/5687",
  "serverity": "\u9ad8",
  "submitTime": "2025-08-13",
  "title": "NVIDIA Triton Inference Server Python\u540e\u7aef\u8d8a\u754c\u5199\u6f0f\u6d1e"
}

CVE-2025-23319 (GCVE-0-2025-23319)

Vulnerability from cvelistv5 – Published: 2025-08-06 12:37 – Updated: 2025-08-06 15:36

Summary

NVIDIA Triton Inference Server for Windows and Linux contains a vulnerability in the Python backend, where an attacker could cause an out-of-bounds write by sending a request. A successful exploit of this vulnerability might lead to remote code execution, denial of service, data tampering, or information disclosure.

Severity

8.1 (High)


                        
                          CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H

SSVC

Exploitation: none Automatable: no Technical Impact: total

CISA Coordinator · CISA-ADP (v2.0.3)

Decision recorded 2025-08-06 14:39 UTC

CWE

CWE-805 - Buffer Access with Incorrect Length Value

Assigner

nvidia CNA under the mitre root

CNA scorecard C 65/100 over 329 records in the last 180 days details

References

3 references

URL	Tags
https://nvd.nist.gov/vuln/detail/CVE-2025-23319
https://www.cve.org/CVERecord?id=CVE-2025-23319
https://nvidia.custhelp.com/app/answers/detail/a_…

Impacted products

1 product

Vendor	Product	Version
NVIDIA	Triton Inference Server	Affected: All versions prior to 25.07

Show details on NVD website

JSON

To clipboard

{
  "containers": {
    "adp": [
      {
        "metrics": [
          {
            "other": {
              "content": {
                "id": "CVE-2025-23319",
                "options": [
                  {
                    "Exploitation": "none"
                  },
                  {
                    "Automatable": "no"
                  },
                  {
                    "Technical Impact": "total"
                  }
                ],
                "role": "CISA Coordinator",
                "timestamp": "2025-08-06T14:39:59.013915Z",
                "version": "2.0.3"
              },
              "type": "ssvc"
            }
          }
        ],
        "providerMetadata": {
          "dateUpdated": "2025-08-06T15:36:16.230Z",
          "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "shortName": "CISA-ADP"
        },
        "title": "CISA ADP Vulnrichment"
      }
    ],
    "cna": {
      "affected": [
        {
          "defaultStatus": "unaffected",
          "platforms": [
            "Windows",
            "Linux"
          ],
          "product": "Triton Inference Server",
          "vendor": "NVIDIA",
          "versions": [
            {
              "status": "affected",
              "version": "All versions prior to 25.07"
            }
          ]
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": true,
              "type": "text/html",
              "value": "NVIDIA Triton Inference Server for Windows and Linux contains a vulnerability in the Python backend, where an attacker could cause an out-of-bounds write by sending a request. A successful exploit of this vulnerability might lead to remote code execution, denial of service, data tampering, or information disclosure."
            }
          ],
          "value": "NVIDIA Triton Inference Server for Windows and Linux contains a vulnerability in the Python backend, where an attacker could cause an out-of-bounds write by sending a request. A successful exploit of this vulnerability might lead to remote code execution, denial of service, data tampering, or information disclosure."
        }
      ],
      "impacts": [
        {
          "descriptions": [
            {
              "lang": "en",
              "value": "Code execution, denial of service, data tampering, information disclosure"
            }
          ]
        }
      ],
      "metrics": [
        {
          "cvssV3_1": {
            "attackComplexity": "HIGH",
            "attackVector": "NETWORK",
            "availabilityImpact": "HIGH",
            "baseScore": 8.1,
            "baseSeverity": "HIGH",
            "confidentialityImpact": "HIGH",
            "integrityImpact": "HIGH",
            "privilegesRequired": "NONE",
            "scope": "UNCHANGED",
            "userInteraction": "NONE",
            "vectorString": "CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H",
            "version": "3.1"
          },
          "format": "CVSS",
          "scenarios": [
            {
              "lang": "en",
              "value": "GENERAL"
            }
          ]
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-805",
              "description": "CWE-805 Buffer Access with Incorrect Length Value",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "dateUpdated": "2025-08-06T12:37:06.672Z",
        "orgId": "9576f279-3576-44b5-a4af-b9a8644b2de6",
        "shortName": "nvidia"
      },
      "references": [
        {
          "url": "https://nvd.nist.gov/vuln/detail/CVE-2025-23319"
        },
        {
          "url": "https://www.cve.org/CVERecord?id=CVE-2025-23319"
        },
        {
          "url": "https://nvidia.custhelp.com/app/answers/detail/a_id/5687"
        }
      ],
      "source": {
        "discovery": "UNKNOWN"
      },
      "x_generator": {
        "engine": "NVIDIA PSIRT"
      }
    }
  },
  "cveMetadata": {
    "assignerOrgId": "9576f279-3576-44b5-a4af-b9a8644b2de6",
    "assignerShortName": "nvidia",
    "cveId": "CVE-2025-23319",
    "datePublished": "2025-08-06T12:37:06.672Z",
    "dateReserved": "2025-01-14T01:06:28.099Z",
    "dateUpdated": "2025-08-06T15:36:16.230Z",
    "state": "PUBLISHED"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.1"
}

Sightings

Author	Source	Type	Date	Other

Nomenclature

Seen: The vulnerability was mentioned, discussed, or observed by the user.
Confirmed: The vulnerability has been validated from an analyst's perspective.
Published Proof of Concept: A public proof of concept is available for this vulnerability.
Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
Not confirmed: The user expressed doubt about the validity of the vulnerability.
Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.

Vulnerability timeline

Sightings and recorded dates, grouped by day · UTC

Order JSON

Publication & reservation Sightings KEV catalogs Other timestamps

Loading timeline…

The MITRE ATT&CK techniques below are AI-generated suggestions, inferred from the description of the vulnerability by the CIRCL/vulnerability-attack-technique-classification-roberta-base model, served locally by ML-Gateway. They have not been verified by an analyst and are provided for guidance only.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.

Related by attack behaviour

Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.

Action not permitted