CVE-2026-94627: vLLM Mooncake connector flaw enables memory exhaustion

Summary

The vLLM Mooncake connector through 0.29.0 does not properly manage GPU KV cache block ownership when concurrent child requests share a single transfer ID in prefill/decode disaggregated deployments. An attacker can submit completion requests with multiple prompts to trigger GPU memory exhaustion, as orphaned KV cache blocks accumulate until the process restarts and legitimate requests can no longer execute.

Mitigation

The source does not state a fix yet. Check the original advisory for updates.

Topics

• Inference infrastructure

Related items

• MediumGHSA-h7c4-q426-7jh9: tsParticles: A single anonymous MCP session can monopolize HTTP mode and deny new clientsSimilar attack · GitHub Advisory Database
• HighGHSA-4p8p-pjpq-2w7h: tsParticles: Malformed JSON requests can terminate the @tsparticles/mcp-server HTTP processSimilar attack · GitHub Advisory Database
• HighGHSA-447x-qqwf-m7xq: tsParticles: Unbounded HTTP Request Body Buffering in @tsparticles/mcp-server Leads to Remote DoSSimilar attack · GitHub Advisory Database
• HighCVE-2026-123356: GROBID unauthenticated thread exhaustion through /api/modelTraining endpointSimilar attack · NVD/CVE Database
• MediumCVE-2026-83494: training-operator immutability check bypass in PyTorchJob ValidateUpdate webhookSimilar attack · NVD/CVE Database