Описание
vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.22.1, the vLLM Dockerfile is vulnerable to a dependency confusion attack through the flashinfer-jit-cache package. The package is installed from a custom index (flashinfer.ai/whl/) using --extra-index-url, but the package name was not registered on PyPI, and UV_INDEX_STRATEGY="unsafe-best-match" is set globally. An attacker who registers flashinfer-jit-cache on PyPI with version 0.6.11.post2 can execute arbitrary code as root during the Docker build and backdoor every resulting container image, enabling exfiltration of all user prompts, API credentials, and model data from production vLLM deployments This vulnerability is fixed in 0.22.1.
A flaw was found in vLLM, an inference and serving engine for large language models (LLMs). This vulnerability, a dependency confusion attack, allows a remote attacker to execute arbitrary code with root privileges during the Docker build process. By exploiting this, an attacker can compromise the resulting container image, leading to the exfiltration of sensitive information like user prompts, API credentials, and model data from production vLLM deployments.
Отчет
CVE-2026-54232 is a build-time dependency confusion issue in upstream vLLM Dockerfiles before 0.22.1. It does not allow remote exploitation of a running vLLM inference service. Red Hat OpenShift AI is not affected. Red Hat AI Inference Server and RHEL AI CUDA images that include flashinfer-jit-cache are in scope for build-process review, but Red Hat has no evidence that shipped images were compromised. Red Hat rates this Moderate for affected products.
Меры по смягчению последствий
Mitigation for this issue is either not available or the currently available options do not meet the Red Hat Product Security criteria comprising ease of use and deployment, applicability to widespread installation base, or stability.
Затронутые пакеты
| Платформа | Пакет | Состояние | Рекомендация | Релиз |
|---|---|---|---|---|
| Red Hat AI Inference Server | rhaiis/vllm-cpu-rhel9 | Not affected | ||
| Red Hat AI Inference Server | rhaiis/vllm-cuda-rhel9 | Affected | ||
| Red Hat AI Inference Server | rhaiis/vllm-neuron-rhel9 | Not affected | ||
| Red Hat AI Inference Server | rhaiis/vllm-rocm-rhel9 | Not affected | ||
| Red Hat AI Inference Server | rhaiis/vllm-spyre-rhel9 | Not affected | ||
| Red Hat AI Inference Server | rhaiis/vllm-tpu-rhel9 | Not affected | ||
| Red Hat AI Inference Server | rhaii/vllm-cpu-rhel9 | Not affected | ||
| Red Hat AI Inference Server | rhaii/vllm-cuda-rhel9 | Affected | ||
| Red Hat AI Inference Server | rhaii/vllm-gaudi-rhel9 | Not affected | ||
| Red Hat AI Inference Server | rhaii/vllm-neuron-rhel9 | Not affected |
Показывать по
Дополнительная информация
Статус:
EPSS
5.7 Medium
CVSS3
Связанные уязвимости
vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.22.1, the vLLM Dockerfile is vulnerable to a dependency confusion attack through the flashinfer-jit-cache package. The package is installed from a custom index (flashinfer.ai/whl/) using --extra-index-url, but the package name was not registered on PyPI, and UV_INDEX_STRATEGY="unsafe-best-match" is set globally. An attacker who registers flashinfer-jit-cache on PyPI with version 0.6.11.post2 can execute arbitrary code as root during the Docker build and backdoor every resulting container image, enabling exfiltration of all user prompts, API credentials, and model data from production vLLM deployments This vulnerability is fixed in 0.22.1.
vLLM is an inference and serving engine for large language models (LLM ...
EPSS
5.7 Medium
CVSS3