{
 "id": "AIR-2026-0236",
 "slug": "vllm-loose-cache-salt-validation-causes-engine-crash-on-lmca",
 "title": "vLLM loose cache_salt validation causes engine crash on LMCache-MP",
 "summary": "Incomplete cache_salt validation in vLLM allows specially crafted requests to raise uncaught exceptions, terminating the EngineCore on certain deployments. A patch has been released in version 0.30.0.",
 "type": "attack_on_ai",
 "categories": [
  "infra_vuln"
 ],
 "severity": 2,
 "targets": [
  "framework_or_tooling"
 ],
 "products": [
  "vLLM"
 ],
 "incident_date": "2026-10-05",
 "incident_date_precision": "day",
 "first_reported": "2026-10-05",
 "sources": [
  {
   "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-105756",
   "publisher": "NVD",
   "title": "CVE-2026-105756: vLLM is an inference and serving engine for large language models.",
   "published_at": "2026-10-05"
  },
  {
   "url": "https://github.com/advisories/GHSA-2823-qmq8-rwvj",
   "publisher": "GitHub Advisory Database",
   "title": "CVE-2026-105756: vLLM: Loose `cache_salt` validation lets a single request kill EngineCore on LMCache-MP deployments — uncaught downstream `ValueError`…",
   "published_at": "2026-10-05"
  }
 ],
 "source_count": 2,
 "confidence": 0.95,
 "status": "published",
 "origin": "seed_nvd",
 "tags": [
  "CVE-2026-105756",
  "denial-of-service",
  "vllm"
 ],
 "created_at": "2026-10-18T06:33:31.967Z",
 "updated_at": "2026-10-18T06:33:31.967Z"
}