Логотип exploitDog
Консоль
Логотип exploitDog

exploitDog

redhat логотип

CVE-2026-57516

Опубликовано: 01 июл. 2026
Источник: redhat
CVSS3: 8.8
EPSS Низкий

Описание

Ray prior to 2.56.0 contains an unsafe deserialization vulnerability in the WebDataset reader that allows attackers to achieve remote code execution by supplying a malicious tar archive to the read_webdataset() function. The _default_decoder() function in webdataset_datasource.py unconditionally calls pickle.loads() on tar entries with .pkl/.pickle extensions and torch.load() with weights_only=False on .pt/.pth entries, executing arbitrary code inside Ray remote workers on every worker that processes the malicious archive.

A flaw was found in Ray. This unsafe deserialization vulnerability in the WebDataset reader allows a remote attacker to achieve arbitrary code execution. By supplying a specially crafted malicious tar archive to the read_webdataset() function, an attacker can trigger the unconditional deserialization of .pkl/.pickle or .pt/.pth entries, leading to the execution of arbitrary code within Ray remote workers.

Отчет

This Important vulnerability in Ray's WebDataset reader allows for remote code execution. An attacker could exploit this by providing a specially crafted tar archive to the read_webdataset() function, leading to arbitrary code execution within Ray remote workers. This risk is present in Red Hat environments where Ray is deployed to process untrusted data archives, as the deserialization of malicious content occurs unconditionally.

Меры по смягчению последствий

To mitigate this issue, restrict the processing of untrusted tar archives by the Ray WebDataset reader. Ensure that only verified and trusted data sources are supplied to the read_webdataset() function. Implement strict access controls and input validation for data ingestion pipelines that interact with Ray's WebDataset reader to prevent the introduction of malicious archives.

Затронутые пакеты

ПлатформаПакетСостояниеРекомендацияРелиз
Red Hat AI Inference Serverrhaii/vllm-gaudi-rhel9Will not fix
Red Hat Enterprise Linux AI (RHEL AI) 3rhelai3/bootc-gaudi-rhel9Affected
Red Hat OpenShift AI (RHOAI)rhoai/odh-kserve-agent-rhel9Not affected
Red Hat OpenShift AI (RHOAI)rhoai/odh-kserve-controller-rhel9Not affected
Red Hat OpenShift AI (RHOAI)rhoai/odh-kserve-router-rhel9Not affected
Red Hat OpenShift AI (RHOAI)rhoai/odh-kserve-storage-initializer-rhel9Not affected
Red Hat OpenShift AI (RHOAI)rhoai/odh-pipeline-runtime-datascience-cpu-py312-rhel9Affected
Red Hat OpenShift AI (RHOAI)rhoai/odh-pipeline-runtime-pytorch-cuda-py312-rhel9Affected
Red Hat OpenShift AI (RHOAI)rhoai/odh-pipeline-runtime-pytorch-rocm-py312-rhel9Affected
Red Hat OpenShift AI (RHOAI)rhoai/odh-pipeline-runtime-tensorflow-cuda-py312-rhel9Affected

Показывать по

Дополнительная информация

Статус:

Important
Дефект:
CWE-502
https://bugzilla.redhat.com/show_bug.cgi?id=2496087ray: Ray: Remote code execution via unsafe deserialization in WebDataset reader

EPSS

Процентиль: 56%
0.00858
Низкий

8.8 High

CVSS3

Связанные уязвимости

CVSS3: 8.8
nvd
2 месяца назад

Ray prior to 2.56.0 contains an unsafe deserialization vulnerability in the WebDataset reader that allows attackers to achieve remote code execution by supplying a malicious tar archive to the read_webdataset() function. The _default_decoder() function in webdataset_datasource.py unconditionally calls pickle.loads() on tar entries with .pkl/.pickle extensions and torch.load() with weights_only=False on .pt/.pth entries, executing arbitrary code inside Ray remote workers on every worker that processes the malicious archive.

CVSS3: 8.8
github
около 1 месяца назад

Ray: Arbitrary code execution via ray.data.read_webdataset default decoder: pickle.loads(value) and torch.load(weights_only=False)

EPSS

Процентиль: 56%
0.00858
Низкий

8.8 High

CVSS3