Vulnerability-Lookup

BDU:2026-03420

Vulnerability from fstec - Published: 03.04.2025

Title

Уязвимость класса PyNcclPipe библиотеки для работы с большими языковыми моделями (LLM) vLLM, позволяющая нарушителю выполнить произвольный код

Description

Уязвимость класса PyNcclPipe библиотеки для работы с большими языковыми моделями (LLM) vLLM связана с недостатками механизма десериализации. Эксплуатация уязвимости может позволить нарушителю, действующему удаленно, выполнить произвольный код

Severity


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

Vendor

vLLM Project

Software Name

vLLM

Software Version

от 0.6.5 до 0.8.5 (vLLM)

Possible Mitigations

Использование рекомендаций производителя: https://github.com/vllm-project/vllm/commit/0d6e187e88874c39cda7409cf673f9e6546893e7

Reference

https://github.com/vllm-project/vllm/security/advisories/GHSA-hjq4-87xh-g4fv https://github.com/vllm-project/vllm/commit/0d6e187e88874c39cda7409cf673f9e6546893e7 https://github.com/vllm-project/vllm/pull/15988 https://docs.vllm.ai/en/latest/deployment/security.html

CWE

CWE-502

JSON

To clipboard

{
  "CVSS 2.0": "AV:N/AC:L/Au:N/C:C/I:C/A:C",
  "CVSS 3.0": "AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
  "CVSS 4.0": null,
  "remediation_\u0418\u0434\u0435\u043d\u0442\u0438\u0444\u0438\u043a\u0430\u0442\u043e\u0440": null,
  "remediation_\u041d\u0430\u0438\u043c\u0435\u043d\u043e\u0432\u0430\u043d\u0438\u0435": null,
  "\u0412\u0435\u043d\u0434\u043e\u0440 \u041f\u041e": "vLLM Project",
  "\u0412\u0435\u0440\u0441\u0438\u044f \u041f\u041e": "\u043e\u0442 0.6.5 \u0434\u043e 0.8.5 (vLLM)",
  "\u0412\u043e\u0437\u043c\u043e\u0436\u043d\u044b\u0435 \u043c\u0435\u0440\u044b \u043f\u043e \u0443\u0441\u0442\u0440\u0430\u043d\u0435\u043d\u0438\u044e": "\u0418\u0441\u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u043d\u0438\u0435 \u0440\u0435\u043a\u043e\u043c\u0435\u043d\u0434\u0430\u0446\u0438\u0439 \u043f\u0440\u043e\u0438\u0437\u0432\u043e\u0434\u0438\u0442\u0435\u043b\u044f:\nhttps://github.com/vllm-project/vllm/commit/0d6e187e88874c39cda7409cf673f9e6546893e7",
  "\u0414\u0430\u0442\u0430 \u0432\u044b\u044f\u0432\u043b\u0435\u043d\u0438\u044f": "03.04.2025",
  "\u0414\u0430\u0442\u0430 \u043f\u043e\u0441\u043b\u0435\u0434\u043d\u0435\u0433\u043e \u043e\u0431\u043d\u043e\u0432\u043b\u0435\u043d\u0438\u044f": "20.03.2026",
  "\u0414\u0430\u0442\u0430 \u043f\u0443\u0431\u043b\u0438\u043a\u0430\u0446\u0438\u0438": "20.03.2026",
  "\u0418\u0434\u0435\u043d\u0442\u0438\u0444\u0438\u043a\u0430\u0442\u043e\u0440": "BDU:2026-03420",
  "\u0418\u0434\u0435\u043d\u0442\u0438\u0444\u0438\u043a\u0430\u0442\u043e\u0440\u044b \u0434\u0440\u0443\u0433\u0438\u0445 \u0441\u0438\u0441\u0442\u0435\u043c \u043e\u043f\u0438\u0441\u0430\u043d\u0438\u0439 \u0443\u044f\u0437\u0432\u0438\u043c\u043e\u0441\u0442\u0438": "CVE-2025-47277",
  "\u0418\u043d\u0444\u043e\u0440\u043c\u0430\u0446\u0438\u044f \u043e\u0431 \u0443\u0441\u0442\u0440\u0430\u043d\u0435\u043d\u0438\u0438": "\u0423\u044f\u0437\u0432\u0438\u043c\u043e\u0441\u0442\u044c \u0443\u0441\u0442\u0440\u0430\u043d\u0435\u043d\u0430",
  "\u041a\u043b\u0430\u0441\u0441 \u0443\u044f\u0437\u0432\u0438\u043c\u043e\u0441\u0442\u0438": "\u0423\u044f\u0437\u0432\u0438\u043c\u043e\u0441\u0442\u044c \u0430\u0440\u0445\u0438\u0442\u0435\u043a\u0442\u0443\u0440\u044b",
  "\u041d\u0430\u0437\u0432\u0430\u043d\u0438\u0435 \u041f\u041e": "vLLM",
  "\u041d\u0430\u0438\u043c\u0435\u043d\u043e\u0432\u0430\u043d\u0438\u0435 \u041e\u0421 \u0438 \u0442\u0438\u043f \u0430\u043f\u043f\u0430\u0440\u0430\u0442\u043d\u043e\u0439 \u043f\u043b\u0430\u0442\u0444\u043e\u0440\u043c\u044b": null,
  "\u041d\u0430\u0438\u043c\u0435\u043d\u043e\u0432\u0430\u043d\u0438\u0435 \u0443\u044f\u0437\u0432\u0438\u043c\u043e\u0441\u0442\u0438": "\u0423\u044f\u0437\u0432\u0438\u043c\u043e\u0441\u0442\u044c \u043a\u043b\u0430\u0441\u0441\u0430 PyNcclPipe \u0431\u0438\u0431\u043b\u0438\u043e\u0442\u0435\u043a\u0438 \u0434\u043b\u044f \u0440\u0430\u0431\u043e\u0442\u044b \u0441 \u0431\u043e\u043b\u044c\u0448\u0438\u043c\u0438 \u044f\u0437\u044b\u043a\u043e\u0432\u044b\u043c\u0438 \u043c\u043e\u0434\u0435\u043b\u044f\u043c\u0438 (LLM) vLLM, \u043f\u043e\u0437\u0432\u043e\u043b\u044f\u044e\u0449\u0430\u044f \u043d\u0430\u0440\u0443\u0448\u0438\u0442\u0435\u043b\u044e \u0432\u044b\u043f\u043e\u043b\u043d\u0438\u0442\u044c \u043f\u0440\u043e\u0438\u0437\u0432\u043e\u043b\u044c\u043d\u044b\u0439 \u043a\u043e\u0434",
  "\u041d\u0430\u043b\u0438\u0447\u0438\u0435 \u044d\u043a\u0441\u043f\u043b\u043e\u0439\u0442\u0430": "\u0421\u0443\u0449\u0435\u0441\u0442\u0432\u0443\u0435\u0442 \u0432 \u043e\u0442\u043a\u0440\u044b\u0442\u043e\u043c \u0434\u043e\u0441\u0442\u0443\u043f\u0435",
  "\u041e\u043f\u0438\u0441\u0430\u043d\u0438\u0435 \u043e\u0448\u0438\u0431\u043a\u0438 CWE": "\u0412\u043e\u0441\u0441\u0442\u0430\u043d\u043e\u0432\u043b\u0435\u043d\u0438\u0435 \u0432 \u043f\u0430\u043c\u044f\u0442\u0438 \u043d\u0435\u0434\u043e\u0441\u0442\u043e\u0432\u0435\u0440\u043d\u044b\u0445 \u0434\u0430\u043d\u043d\u044b\u0445 (CWE-502)",
  "\u041e\u043f\u0438\u0441\u0430\u043d\u0438\u0435 \u0443\u044f\u0437\u0432\u0438\u043c\u043e\u0441\u0442\u0438": "\u0423\u044f\u0437\u0432\u0438\u043c\u043e\u0441\u0442\u044c \u043a\u043b\u0430\u0441\u0441\u0430 PyNcclPipe \u0431\u0438\u0431\u043b\u0438\u043e\u0442\u0435\u043a\u0438 \u0434\u043b\u044f \u0440\u0430\u0431\u043e\u0442\u044b \u0441 \u0431\u043e\u043b\u044c\u0448\u0438\u043c\u0438 \u044f\u0437\u044b\u043a\u043e\u0432\u044b\u043c\u0438 \u043c\u043e\u0434\u0435\u043b\u044f\u043c\u0438 (LLM) vLLM \u0441\u0432\u044f\u0437\u0430\u043d\u0430 \u0441 \u043d\u0435\u0434\u043e\u0441\u0442\u0430\u0442\u043a\u0430\u043c\u0438 \u043c\u0435\u0445\u0430\u043d\u0438\u0437\u043c\u0430 \u0434\u0435\u0441\u0435\u0440\u0438\u0430\u043b\u0438\u0437\u0430\u0446\u0438\u0438. \u042d\u043a\u0441\u043f\u043b\u0443\u0430\u0442\u0430\u0446\u0438\u044f \u0443\u044f\u0437\u0432\u0438\u043c\u043e\u0441\u0442\u0438 \u043c\u043e\u0436\u0435\u0442 \u043f\u043e\u0437\u0432\u043e\u043b\u0438\u0442\u044c \u043d\u0430\u0440\u0443\u0448\u0438\u0442\u0435\u043b\u044e, \u0434\u0435\u0439\u0441\u0442\u0432\u0443\u044e\u0449\u0435\u043c\u0443 \u0443\u0434\u0430\u043b\u0435\u043d\u043d\u043e, \u0432\u044b\u043f\u043e\u043b\u043d\u0438\u0442\u044c \u043f\u0440\u043e\u0438\u0437\u0432\u043e\u043b\u044c\u043d\u044b\u0439 \u043a\u043e\u0434",
  "\u041f\u043e\u0441\u043b\u0435\u0434\u0441\u0442\u0432\u0438\u044f \u044d\u043a\u0441\u043f\u043b\u0443\u0430\u0442\u0430\u0446\u0438\u0438 \u0443\u044f\u0437\u0432\u0438\u043c\u043e\u0441\u0442\u0438": null,
  "\u041f\u0440\u043e\u0447\u0430\u044f \u0438\u043d\u0444\u043e\u0440\u043c\u0430\u0446\u0438\u044f": null,
  "\u0421\u0432\u044f\u0437\u044c \u0441 \u0438\u043d\u0446\u0438\u0434\u0435\u043d\u0442\u0430\u043c\u0438 \u0418\u0411": "\u0414\u0430\u043d\u043d\u044b\u0435 \u0443\u0442\u043e\u0447\u043d\u044f\u044e\u0442\u0441\u044f",
  "\u0421\u043e\u0441\u0442\u043e\u044f\u043d\u0438\u0435 \u0443\u044f\u0437\u0432\u0438\u043c\u043e\u0441\u0442\u0438": "\u041e\u043f\u0443\u0431\u043b\u0438\u043a\u043e\u0432\u0430\u043d\u0430",
  "\u0421\u043f\u043e\u0441\u043e\u0431 \u0443\u0441\u0442\u0440\u0430\u043d\u0435\u043d\u0438\u044f": "\u041e\u0431\u043d\u043e\u0432\u043b\u0435\u043d\u0438\u0435 \u043f\u0440\u043e\u0433\u0440\u0430\u043c\u043c\u043d\u043e\u0433\u043e \u043e\u0431\u0435\u0441\u043f\u0435\u0447\u0435\u043d\u0438\u044f",
  "\u0421\u043f\u043e\u0441\u043e\u0431 \u044d\u043a\u0441\u043f\u043b\u0443\u0430\u0442\u0430\u0446\u0438\u0438": "\u041c\u0430\u043d\u0438\u043f\u0443\u043b\u0438\u0440\u043e\u0432\u0430\u043d\u0438\u0435 \u0441\u0442\u0440\u0443\u043a\u0442\u0443\u0440\u0430\u043c\u0438 \u0434\u0430\u043d\u043d\u044b\u0445",
  "\u0421\u0441\u044b\u043b\u043a\u0438 \u043d\u0430 \u0438\u0441\u0442\u043e\u0447\u043d\u0438\u043a\u0438": "https://github.com/vllm-project/vllm/security/advisories/GHSA-hjq4-87xh-g4fv\nhttps://github.com/vllm-project/vllm/commit/0d6e187e88874c39cda7409cf673f9e6546893e7\nhttps://github.com/vllm-project/vllm/pull/15988\nhttps://docs.vllm.ai/en/latest/deployment/security.html",
  "\u0421\u0442\u0430\u0442\u0443\u0441 \u0443\u044f\u0437\u0432\u0438\u043c\u043e\u0441\u0442\u0438": "\u041f\u043e\u0434\u0442\u0432\u0435\u0440\u0436\u0434\u0435\u043d\u0430 \u043f\u0440\u043e\u0438\u0437\u0432\u043e\u0434\u0438\u0442\u0435\u043b\u0435\u043c",
  "\u0422\u0438\u043f \u041f\u041e": "\u041f\u041e \u0434\u043b\u044f \u0440\u0430\u0437\u0440\u0430\u0431\u043e\u0442\u043a\u0438 \u0418\u0418",
  "\u0422\u0438\u043f \u043e\u0448\u0438\u0431\u043a\u0438 CWE": "CWE-502",
  "\u0423\u0440\u043e\u0432\u0435\u043d\u044c \u043e\u043f\u0430\u0441\u043d\u043e\u0441\u0442\u0438 \u0443\u044f\u0437\u0432\u0438\u043c\u043e\u0441\u0442\u0438": "\u041a\u0440\u0438\u0442\u0438\u0447\u0435\u0441\u043a\u0438\u0439 \u0443\u0440\u043e\u0432\u0435\u043d\u044c \u043e\u043f\u0430\u0441\u043d\u043e\u0441\u0442\u0438 (\u0431\u0430\u0437\u043e\u0432\u0430\u044f \u043e\u0446\u0435\u043d\u043a\u0430 CVSS 2.0 \u0441\u043e\u0441\u0442\u0430\u0432\u043b\u044f\u0435\u0442 10)\n\u041a\u0440\u0438\u0442\u0438\u0447\u0435\u0441\u043a\u0438\u0439 \u0443\u0440\u043e\u0432\u0435\u043d\u044c \u043e\u043f\u0430\u0441\u043d\u043e\u0441\u0442\u0438 (\u0431\u0430\u0437\u043e\u0432\u0430\u044f \u043e\u0446\u0435\u043d\u043a\u0430 CVSS 3.1 \u0441\u043e\u0441\u0442\u0430\u0432\u043b\u044f\u0435\u0442 9,8)"
}

CVE-2025-47277 (GCVE-0-2025-47277)

Vulnerability from cvelistv5 – Published: 2025-05-20 17:32 – Updated: 2025-05-20 17:52

Title

vLLM Allows Remote Code Execution via PyNcclPipe Communication Service

Summary

vLLM, an inference and serving engine for large language models (LLMs), has an issue in versions 0.6.5 through 0.8.4 that ONLY impacts environments using the `PyNcclPipe` KV cache transfer integration with the V0 engine. No other configurations are affected. vLLM supports the use of the `PyNcclPipe` class to establish a peer-to-peer communication domain for data transmission between distributed nodes. The GPU-side KV-Cache transmission is implemented through the `PyNcclCommunicator` class, while CPU-side control message passing is handled via the `send_obj` and `recv_obj` methods on the CPU side. The intention was that this interface should only be exposed to a private network using the IP address specified by the `--kv-ip` CLI parameter. The vLLM documentation covers how this must be limited to a secured network. The default and intentional behavior from PyTorch is that the `TCPStore` interface listens on ALL interfaces, regardless of what IP address is provided. The IP address given was only used as a client-side address to use. vLLM was fixed to use a workaround to force the `TCPStore` instance to bind its socket to a specified private interface. As of version 0.8.5, vLLM limits the `TCPStore` socket to the private interface as configured.

Severity

9.8 (Critical)


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

SSVC

Exploitation: none Automatable: yes Technical Impact: total

CISA Coordinator · CISA-ADP (v2.0.3)

Decision recorded 2025-05-20 17:52 UTC

CWE

CWE-502 - Deserialization of Untrusted Data

Assigner

GitHub_M mitre root

References

4 references

URL	Tags
https://github.com/vllm-project/vllm/security/adv…	x_refsource_CONFIRM
https://github.com/vllm-project/vllm/pull/15988	x_refsource_MISC
https://github.com/vllm-project/vllm/commit/0d6e1…	x_refsource_MISC
https://docs.vllm.ai/en/latest/deployment/security.html	x_refsource_MISC

Impacted products

1 product

Vendor	Product	Version
vllm-project	vllm	Affected: >= 0.6.5, < 0.8.5

Show details on NVD website

JSON

To clipboard

{
  "containers": {
    "adp": [
      {
        "metrics": [
          {
            "other": {
              "content": {
                "id": "CVE-2025-47277",
                "options": [
                  {
                    "Exploitation": "none"
                  },
                  {
                    "Automatable": "yes"
                  },
                  {
                    "Technical Impact": "total"
                  }
                ],
                "role": "CISA Coordinator",
                "timestamp": "2025-05-20T17:52:22.643444Z",
                "version": "2.0.3"
              },
              "type": "ssvc"
            }
          }
        ],
        "providerMetadata": {
          "dateUpdated": "2025-05-20T17:52:31.274Z",
          "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "shortName": "CISA-ADP"
        },
        "title": "CISA ADP Vulnrichment"
      }
    ],
    "cna": {
      "affected": [
        {
          "product": "vllm",
          "vendor": "vllm-project",
          "versions": [
            {
              "status": "affected",
              "version": "\u003e= 0.6.5, \u003c 0.8.5"
            }
          ]
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "value": "vLLM, an inference and serving engine for large language models (LLMs), has an issue in versions 0.6.5 through 0.8.4 that ONLY impacts environments using the `PyNcclPipe` KV cache transfer integration with the V0 engine. No other configurations are affected. vLLM supports the use of the\u00a0`PyNcclPipe`\u00a0class to establish a peer-to-peer communication domain for data transmission between distributed nodes. The GPU-side KV-Cache transmission is implemented through the\u00a0`PyNcclCommunicator`\u00a0class, while CPU-side control message passing is handled via the\u00a0`send_obj`\u00a0and\u00a0`recv_obj`\u00a0methods on the CPU side.\u200b The intention was that this interface should only be exposed to a private network using the IP address specified by the `--kv-ip` CLI parameter. The vLLM documentation covers how this must be limited to a secured network. The default and intentional behavior from PyTorch is that the `TCPStore` interface listens on ALL interfaces, regardless of what IP address is provided. The IP address given was only used as a client-side address to use. vLLM was fixed to use a workaround to force the `TCPStore` instance to bind its socket to a specified private interface. As of version 0.8.5, vLLM limits the `TCPStore` socket to the private interface as configured."
        }
      ],
      "metrics": [
        {
          "cvssV3_1": {
            "attackComplexity": "LOW",
            "attackVector": "NETWORK",
            "availabilityImpact": "HIGH",
            "baseScore": 9.8,
            "baseSeverity": "CRITICAL",
            "confidentialityImpact": "HIGH",
            "integrityImpact": "HIGH",
            "privilegesRequired": "NONE",
            "scope": "UNCHANGED",
            "userInteraction": "NONE",
            "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
            "version": "3.1"
          }
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-502",
              "description": "CWE-502: Deserialization of Untrusted Data",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "dateUpdated": "2025-05-20T17:32:27.034Z",
        "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "shortName": "GitHub_M"
      },
      "references": [
        {
          "name": "https://github.com/vllm-project/vllm/security/advisories/GHSA-hjq4-87xh-g4fv",
          "tags": [
            "x_refsource_CONFIRM"
          ],
          "url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-hjq4-87xh-g4fv"
        },
        {
          "name": "https://github.com/vllm-project/vllm/pull/15988",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/vllm-project/vllm/pull/15988"
        },
        {
          "name": "https://github.com/vllm-project/vllm/commit/0d6e187e88874c39cda7409cf673f9e6546893e7",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/vllm-project/vllm/commit/0d6e187e88874c39cda7409cf673f9e6546893e7"
        },
        {
          "name": "https://docs.vllm.ai/en/latest/deployment/security.html",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://docs.vllm.ai/en/latest/deployment/security.html"
        }
      ],
      "source": {
        "advisory": "GHSA-hjq4-87xh-g4fv",
        "discovery": "UNKNOWN"
      },
      "title": "vLLM Allows Remote Code Execution via PyNcclPipe Communication Service"
    }
  },
  "cveMetadata": {
    "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
    "assignerShortName": "GitHub_M",
    "cveId": "CVE-2025-47277",
    "datePublished": "2025-05-20T17:32:27.034Z",
    "dateReserved": "2025-05-05T16:53:10.373Z",
    "dateUpdated": "2025-05-20T17:52:31.274Z",
    "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.

Detection rules are retrieved from Rulezet.

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.