CVE-2026-94624: vllm p2p never connected zmq_exhaustion
When the OffloadingConnector uses TieringOffloadingSpec with a P2P secondary tier, an API client can submit many unique non-listening peer addresses through kv_transfer_params.prefill. vLLM keeps every never-established P2P session and its ZeroMQ sockets after the request finishes. On vLLM 0.26.0, 340 such peers filled the default ZeroMQ context socket quota; creating the next peer raised an uncaught ZMQError, terminated EngineCore, and shut down the API server. The path was reproduced with a normal API key and no access to the internal P2P listener.
References
- github.com/advisories/GHSA-frvg-qw88-hww9
- github.com/pypa/advisory-database/tree/main/vulns/vllm/PYSEC-2026-4006.yaml
- github.com/vllm-project/vllm/blob/v0.29.0/vllm/v1/kv_offload/tiering/p2p/control/zmq.py
- github.com/vllm-project/vllm/blob/v0.29.0/vllm/v1/kv_offload/tiering/p2p/manager.py
- github.com/vllm-project/vllm/commit/93d2cab35e8ae66bbaa1dea1c077986d5a79eb80
- github.com/vllm-project/vllm/pull/51504
- github.com/vllm-project/vllm/releases/tag/v0.32.0
- github.com/vllm-project/vllm/security/advisories/GHSA-frvg-qw88-hww9
- nvd.nist.gov/vuln/detail/CVE-2026-94624
- www.vulncheck.com/advisories/vllm-through-0.29.0-denial-of-service-via-unbounded-p2p-kv-offloading-sessions
Code Behaviors & Features
Detect and mitigate CVE-2026-94624 with GitLab Dependency Scanning
Secure your software supply chain by verifying that all open source dependencies used in your projects contain no disclosed vulnerabilities. Dependency Scanning using SBOM also detects vulnerabilities from proprietary advisories that this site does not include. Learn more about Dependency Scanning using SBOM →