{"id":"ghsa-ghsa-frvg-qw88-hww9","seq":1708821,"kind":"advisory","title":"vllm p2p never connected zmq_exhaustion","url":"https://github.com/advisories/GHSA-frvg-qw88-hww9","permalink":"https://vulnpulse.app/items/ghsa-ghsa-frvg-qw88-hww9","summary":"影響: pip:vllm < 0.32.0 → 0.32.0\nWhen 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 n…","severity":"medium","techs":["python"],"cves":["CVE-2026-94624"],"tags":["GHSA-frvg-qw88-hww9","vllm"],"source":{"id":"ghsa","name":"GitHub Advisory Database"},"published_at":"2026-10-27T14:45:53.000Z","fetched_at":"2026-10-27T21:03:52.214Z"}