RHSA-2026:36005HighCVSS 9.1

Red Hat Security Advisory: Red Hat AI Inference Server 3.2.2 (CUDA)

Published
July 6, 2026
Last Modified
August 17, 2026

🔗 CVE IDs covered (12)

CVE-2026-26740 →CVE-2026-33846 →CVE-2026-34755 →CVE-2026-34756 →CVE-2026-34982 →CVE-2026-48746 →CVE-2026-54235 →CVE-2026-33845 →CVE-2026-41523 →CVE-2026-42009 →CVE-2026-42010 →CVE-2026-54283 →

📋 Description

CVE-2026-26740 — giflib: giflib: Denial of Service via buffer overflow in EGifGCBToExtension CVE-2026-33845 — gnutls: GnuTLS: Denial of Service via DTLS zero-length fragment CVE-2026-33846 — gnutls: GnuTLS: Denial of Service via heap buffer overflow in DTLS handshake fragment reassembly CVE-2026-34755 — vLLM: vLLM: Denial of Service due to excessive video frame processing CVE-2026-34756 — vllm: vLLM: Denial of Service via excessively large 'n' parameter in OpenAI-compatible API CVE-2026-34982 — vim: arbitrary command execution via modeline sandbox bypass CVE-2026-41523 — vllm: vLLM: Arbitrary code execution via malicious HuggingFace model CVE-2026-42009 — gnutls: gnutls: Denial of Service via DTLS packet reordering vulnerability CVE-2026-42010 — gnutls: gnutls: Authentication Bypass via NUL Character in Username CVE-2026-48746 — vllm: starlette: vLLM: Critical authentication bypass allows unauthorized API access CVE-2026-54235 — vllm: vLLM: Denial of Service due to improper floating-point validation CVE-2026-54283 — starlette: Starlette: request.form() limits silently ignored for application/x-www-form-urlencoded enable DoS

🎯 Affected products3

  • Red Hat AI Inference Server 3.2
  • registry.redhat.io/rhaiis/vllm-cuda-rhel9@sha256:6b93986199cc8e02e9672374938a2aac7b7452d1bf7d1a8a6445bfe1940f6328_amd64 as a component of Red Hat AI Inference Server 3.2
  • registry.redhat.io/rhaiis/vllm-cuda-rhel9@sha256:9533230dd9b2ab392c96f4e8a556ddbebf0be308befdf92691dc7e228da361bf_arm64 as a component of Red Hat AI Inference Server 3.2

✅ Remediation

For more information visit https://access.redhat.com/errata/RHSA-2026:36005 Workaround: 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. Workaround: 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. Workaround: To mitigate this issue, disable the modeline support by adding the following command to the Vim configuration file: ~~~ set nomodeline ~~~ Workaround: Avoid running vLLM with python -O or PYTHONOPTIMIZE=1 until updated packages are available. Only load models from trusted sources. Restrict who can deploy or update models on inference endpoints. Apply network access controls and authentication in front of vLLM APIs. Workaround: Restrict network access to the vLLM API endpoint to only trusted clients and internal networks. Implement firewall rules or network policies to limit inbound connections to the vLLM service, thereby reducing the attack surface. This operational control helps prevent unauthorized external access to the vulnerable API.

🔗 References (16)