vllm: vllm: Information disclosure via SSRF in X-Request-Id header (CVE-2026-116176)

Overview

A flaw was found in vLLM. This vulnerability allows a remote attacker to trigger a Server-Side Request Forgery (SSRF) by submitting an HTTP request containing a crafted X-Request-Id header. Because the application extracts network addresses from this header without validation, backend workers can be forced to initiate outbound network connections to an attacker-controlled server. This flaw can lead to severe information disclosure by leaking sensitive model cache data and internal identifiers, or result in a Denial of Service (DoS) by stalling worker processes.

Vendor advisories

• Red Hat VEX · Important · affected: Red Hat AI Inference Server, Red Hat Enterprise Linux AI (RHEL AI) 3, Red Hat OpenShift AI (RHOAI) · no fix planned: Red Hat AI Inference Server, Red Hat Enterprise Linux AI (RHEL AI) 3, Red Hat OpenShift AI (RHOAI) · updated 2026-10-26 · vex

vllm: vllm: Information disclosure via SSRF in X-Request-Id header — rated Important by Red Hat. Released 2026-10-25, updated 2026-10-26.

Affected:

• Red Hat AI Inference Server
• Red Hat Enterprise Linux AI (RHEL AI) 3
• Red Hat OpenShift AI (RHOAI)

No fix planned:

• Red Hat AI Inference Server
• Red Hat Enterprise Linux AI (RHEL AI) 3
• Red Hat OpenShift AI (RHOAI)

Not affected:

• Red Hat OpenShift AI (RHOAI)

Remediation

Affected