High [CVE-2026-94622] Denial of Service via Incomplete NIXL KV Transfer Metadata
This high-severity Red Hat Linux advisory covers CVE-2026-94622; related products: Red Hat AI Inference Server, Red Hat Enterprise Linux AI (RHEL AI) 3, Red Hat OpenShift AI (RHOAI).
Aggregated and source-linked by VulniPulse. Data sources, validation and limitations.
VulniPulse record published Record updated
Android app · Google Play
Monitor future Red Hat Linux CVEs from your phone.
Choose a whole vendor or a precise platform, then receive matching security advisories by phone notification, email, or both. Coverage follows 32 official vendor sources and 160+ reviewed platform categories.
Summary
Denial of Service via Incomplete NIXL KV Transfer Metadata. Red Hat rates this important (CVSS 7.5).
Weakness: CWE-248.
Affected products named by the advisory: Red Hat AI Inference Server; Red Hat Enterprise Linux AI (RHEL AI) 3; Red Hat OpenShift AI (RHOAI).
- 0.29.0
Official advisory · high-confidence parse· fetched 8 days ago·verify at source
Fixed versions
No fixed release is recorded yet. That does not prove no patch exists — confirm against the vendor advisory.
Official advisory · high-confidence parse· fetched 8 days ago·verify at source
Mitigation checklist
- To reduce exposure to this vulnerability, apply the following operational controls: 1. Disable disaggregated KV transfer: Deploy vLLM instances in a standalone configuration or run unified prefill and decode workers rather than disaggregated roles utilizing the NIXL KV cache transfer connector. 2. Restrict network exposure: Apply OpenShift NetworkPolicies or firewall rules to limit access to vLLM inference ports, permitting traffic only from trusted API gateways or internal clients. Caveats: Disabling disaggregated prefill/decode operation prevents multi-node distributed cache sharing and may reduce inference performance for long-context requests. Network restrictions will prevent access from non-whitelisted IP addresses. Warning: Altering the deployment architecture requires restarting vLLM pods or containers, which will cause a temporary service disruption during workload restart.
Official advisory · high-confidence parse· fetched 8 days ago·verify at source
Discussion(0)
No comments yet. Share field notes, upgrade gotchas, or questions — verify against the vendor advisory before acting on community advice.
Sign in to join the discussion.