{"id":"CVE-2026-94627","aliases":[],"summary":"vLLM through 0.29.0 GPU KV Cache Leak via Mooncake Transfer ID Collision","published_at":"2026-09-21T22:04:16.559Z","modified_at":"2026-09-25T15:51:42Z","products":[],"rating":{"severity":"high","score":8.7,"vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N","attack_path":"internet_unauthenticated","user_interaction":false,"malicious":false,"red_hat":"Important","vendor_ratings":[{"source":"Red Hat","rating":"Important"}]},"fix_available":false,"kev":{"listed":false,"ransomware":false},"public_exploit":false,"exploits":[],"epss":{"score":0.0062699997797608376,"percentile":0.5085099935531616,"date":"2026-11-19T00:00:00Z"},"priority":10.7,"details":"vLLM Mooncake connector through 0.29.0 fails to properly manage GPU KV cache block ownership when concurrent child requests share a single transfer ID in prefill/decode disaggregated deployments. Attackers can trigger GPU memory exhaustion by submitting completion requests with multiple prompts, causing orphaned KV cache blocks to accumulate until process restart and eventually preventing legitimate requests from executing.","affected_stack":[],"affected_other":[{"ecosystem":"CVE","package":"vllm-project/vllm","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"last_affected":"0.29.0"}]}],"provider":"cvelistv5","source":"CVE-2026-94627","relation":"self"},{"ecosystem":"NVD","package":"vllm:vllm","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"last_affected":"0.29.0"}]}],"provider":"nvd","source":"CVE-2026-94627","relation":"self"}],"product_status":[{"vendor":"Red Hat","product":"Red Hat AI Inference Server","cpe":"cpe:/a:redhat:ai_inference_server:3","status":"affected","packages":["rhaii/vllm-cpu-rhel9","rhaii/vllm-cuda-rhel9","rhaii/vllm-rocm-rhel9","rhaii/vllm-spyre-rhel9"],"package_count":4,"fixed":[],"advisories":[],"note":"Affected","provider":"redhat-vex","source":"CVE-2026-94627"},{"vendor":"Red Hat","product":"Red Hat AI Inference Server","cpe":"cpe:/a:redhat:ai_inference_server:3","status":"affected","packages":["rhaii/vllm-gaudi-rhel9","rhaii/vllm-neuron-rhel9","rhaii/vllm-tpu-rhel9","rhaiis/vllm-cpu-rhel9","rhaiis/vllm-cuda-rhel9","rhaiis/vllm-neuron-rhel9","rhaiis/vllm-rocm-rhel9","rhaiis/vllm-spyre-rhel9","rhaiis/vllm-tpu-rhel9"],"package_count":9,"fixed":[],"advisories":[],"note":"Will not fix","provider":"redhat-vex","source":"CVE-2026-94627"},{"vendor":"Red Hat","product":"Red Hat Enterprise Linux AI (RHEL AI) 3","cpe":"cpe:/a:redhat:enterprise_linux_ai:3","status":"affected","packages":["rhelai3/bootc-aws-cuda-rhel9","rhelai3/bootc-azure-cuda-rhel9","rhelai3/bootc-azure-rocm-rhel9","rhelai3/bootc-cuda-rhel9","rhelai3/bootc-gcp-cuda-rhel9","rhelai3/bootc-rocm-rhel9"],"package_count":6,"fixed":[],"advisories":[],"note":"Affected","provider":"redhat-vex","source":"CVE-2026-94627"},{"vendor":"Red Hat","product":"Red Hat Enterprise Linux AI (RHEL AI) 3","cpe":"cpe:/a:redhat:enterprise_linux_ai:3","status":"affected","packages":["rhelai3/bootc-gaudi-rhel9"],"package_count":1,"fixed":[],"advisories":[],"note":"Will not fix","provider":"redhat-vex","source":"CVE-2026-94627"},{"vendor":"Red Hat","product":"Red Hat OpenShift AI (RHOAI)","cpe":"cpe:/a:redhat:openshift_ai","status":"affected","packages":["rhoai/odh-llm-d-kv-cache-rhel9"],"package_count":1,"fixed":[],"advisories":[],"note":"Affected","provider":"redhat-vex","source":"CVE-2026-94627"},{"vendor":"Red Hat","product":"Red Hat OpenShift AI (RHOAI)","cpe":"cpe:/a:redhat:openshift_ai","status":"affected","packages":["rhoai/odh-vllm-gaudi-rhel9"],"package_count":1,"fixed":[],"advisories":[],"note":"Will not fix","provider":"redhat-vex","source":"CVE-2026-94627"},{"vendor":"Red Hat","product":"Red Hat OpenShift AI (RHOAI)","cpe":"cpe:/a:redhat:openshift_ai","status":"not_affected","packages":["rhoai/odh-th-torch-cuda-py312-rhel9"],"package_count":1,"fixed":[],"advisories":[],"note":"vulnerable_code_not_in_execute_path","provider":"redhat-vex","source":"CVE-2026-94627"},{"vendor":"Red Hat","product":"Red Hat OpenShift AI 3.4","cpe":"cpe:/a:redhat:openshift_ai:3.4::el9","status":"fixed","packages":["rhoai/odh-kserve-agent-rhel9","rhoai/odh-kserve-controller-rhel9","rhoai/odh-kserve-router-rhel9","rhoai/odh-kserve-storage-initializer-rhel9"],"package_count":4,"fixed":[],"advisories":["RHSA-2026:84734"],"provider":"redhat-vex","source":"CVE-2026-94627"},{"vendor":"Red Hat","product":"Red Hat OpenShift AI 3.4","cpe":"cpe:/a:redhat:openshift_ai:3.4::el9","status":"not_affected","packages":["rhoai/odh-ai-gateway-payload-processing-rhel9","rhoai/odh-automl-rhel9","rhoai/odh-autorag-rhel9","rhoai/odh-built-in-detector-rhel9","rhoai/odh-cli-rhel9","rhoai/odh-dashboard-rhel9","rhoai/odh-data-science-pipelines-argo-argoexec-rhel9","rhoai/odh-data-science-pipelines-argo-workflowcontroller-rhel9","rhoai/odh-data-science-pipelines-operator-controller-rhel9","rhoai/odh-eval-hub-rhel9","rhoai/odh-feast-operator-rhel9","rhoai/odh-feature-server-rhel9","rhoai/odh-fms-guardrails-orchestrator-rhel9","rhoai/odh-guardrails-detector-huggingface-runtime-rhel9","rhoai/odh-kf-notebook-controller-rhel9","rhoai/odh-kserve-autogluon-server-rhel9","rhoai/odh-kserve-llmisvc-controller-rhel9","rhoai/odh-kube-auth-proxy-rhel9","rhoai/odh-kuberay-operator-controller-rhel9","rhoai/odh-llama-stack-core-rhel9","rhoai/odh-llama-stack-k8s-operator-rhel9","rhoai/odh-llm-d-batch-gateway-apiserver-rhel9","rhoai/odh-llm-d-batch-gateway-gc-rhel9","rhoai/odh-llm-d-batch-gateway-processor-rhel9","rhoai/odh-llm-d-inference-scheduler-rhel9"],"package_count":102,"fixed":[],"advisories":[],"note":"vulnerable_code_not_present","provider":"redhat-vex","source":"CVE-2026-94627"}],"records":[{"provider":"csaf-bsi","source_id":"WID-SEC-W-2026-3507","relation":"upstream","url":"https://wid.cert-bund.de/portal/wid/securityadvisory?name=WID-SEC-2026-3507","license":"TLP:CLEAR, mirroring permitted","published_at":"2026-09-21T22:00:00Z","modified_at":"2026-11-09T23:00:00Z","summary":"vllm: Mehrere Schwachstellen ermöglichen Denial of Service","details":"Ein entfernter, anonymer Angreifer kann mehrere Schwachstellen in vllm ausnutzen, um einen Denial of Service Angriff durchzuführen.\n\nOpen Source vLLM ist eine Open-Source-Bibliothek für schnelle und effiziente Inferenz von Large Language Models (LLMs).","severity":[{"type":"BSI","score":"mittel","source":"WID-SEC-W-2026-3507"}],"affected":[],"product_status":[],"references":[{"type":"WEB","url":"https://wid.cert-bund.de/.well-known/csaf/white/2026/wid-sec-w-2026-3507.json"},{"type":"WEB","url":"https://wid.cert-bund.de/portal/wid/securityadvisory?name=WID-SEC-2026-3507"},{"type":"WEB","url":"https://github.com/advisories/GHSA-4h5x-w852-6g2x"},{"type":"WEB","url":"https://github.com/advisories/GHSA-g3cp-v4cp-9xfq"},{"type":"WEB","url":"https://github.com/advisories/GHSA-8qpc-6rxx-4pxf"},{"type":"WEB","url":"https://github.com/advisories/GHSA-9p52-m77h-4wq5"},{"type":"WEB","url":"https://github.com/advisories/GHSA-hq63-vm56-2666"},{"type":"WEB","url":"https://github.com/advisories/GHSA-pjfg-w894-9457"},{"type":"WEB","url":"https://access.redhat.com/errata/RHSA-2026:94717"},{"type":"WEB","url":"https://access.redhat.com/errata/RHSA-2026:94738"},{"type":"WEB","url":"https://access.redhat.com/errata/RHSA-2026:94709"},{"type":"WEB","url":"https://access.redhat.com/errata/RHSA-2026:94721"},{"type":"WEB","url":"https://access.redhat.com/errata/RHSA-2026:94702"},{"type":"WEB","url":"https://access.redhat.com/errata/RHSA-2026:96333"},{"type":"WEB","url":"https://access.redhat.com/errata/RHSA-2026:96323"}]},{"provider":"cvelistv5","source_id":"CVE-2026-94627","relation":"self","url":"https://www.cve.org/CVERecord?id=CVE-2026-94627","license":"CC-BY-4.0","published_at":"2026-09-21T22:04:16.559Z","modified_at":"2026-09-22T12:51:23.062Z","summary":"vLLM through 0.29.0 GPU KV Cache Leak via Mooncake Transfer ID Collision","details":"vLLM Mooncake connector through 0.29.0 fails to properly manage GPU KV cache block ownership when concurrent child requests share a single transfer ID in prefill/decode disaggregated deployments. Attackers can trigger GPU memory exhaustion by submitting completion requests with multiple prompts, causing orphaned KV cache blocks to accumulate until process restart and eventually preventing legitimate requests from executing.","severity":[{"type":"CVSS_V4","score":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N","source":"CVE-2026-94627"},{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","source":"CVE-2026-94627"}],"affected":[{"ecosystem":"CVE","package":"vllm-project/vllm","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"last_affected":"0.29.0"}]}],"provider":"cvelistv5","source":"CVE-2026-94627","relation":"self"}],"product_status":[],"references":[{"type":"FIX","url":"https://github.com/vllm-project/vllm/pull/49796"},{"type":"WEB","url":"https://github.com/vllm-project/vllm/blob/v0.29.0/vllm/distributed/kv_transfer/kv_connector/v1/mooncake/mooncake_connector.py#L1978-L1989"},{"type":"WEB","url":"https://github.com/vllm-project/vllm"},{"type":"WEB","url":"https://www.vulncheck.com/advisories/vllm-through-0.29.0-gpu-kv-cache-leak-via-mooncake-transfer-id-collision"}]},{"provider":"nvd","source_id":"CVE-2026-94627","relation":"self","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-94627","severity":[],"affected":[{"ecosystem":"NVD","package":"vllm:vllm","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"last_affected":"0.29.0"}]}],"provider":"nvd","source":"CVE-2026-94627","relation":"self"}],"product_status":[],"references":[]},{"provider":"redhat-vex","source_id":"CVE-2026-94627","relation":"self","url":"https://access.redhat.com/security/cve/CVE-2026-94627","license":"CC-BY-4.0","published_at":"2026-09-21T22:04:16.559Z","modified_at":"2026-09-25T15:51:42Z","summary":"vllm: vLLM: Denial of Service via GPU memory exhaustion","details":"A flaw was found in vLLM. A remote attacker can exploit a vulnerability in the Mooncake connector's management of GPU Key-Value (KV) cache block ownership. By submitting completion requests with multiple prompts that share a single transfer ID, attackers can cause orphaned KV cache blocks to accumulate. This leads to GPU memory exhaustion, ultimately preventing legitimate requests from being processed and resulting in a Denial of Service (DoS).","severity":[{"type":"CVSS_V3","score":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H","source":"CVE-2026-94627"},{"type":"Red Hat","score":"Important","source":"CVE-2026-94627"}],"affected":[],"product_status":[{"vendor":"Red Hat","product":"Red Hat AI Inference Server","cpe":"cpe:/a:redhat:ai_inference_server:3","status":"affected","packages":["rhaii/vllm-cpu-rhel9","rhaii/vllm-cuda-rhel9","rhaii/vllm-rocm-rhel9","rhaii/vllm-spyre-rhel9"],"package_count":4,"fixed":[],"advisories":[],"note":"Affected","provider":"redhat-vex","source":"CVE-2026-94627"},{"vendor":"Red Hat","product":"Red Hat AI Inference Server","cpe":"cpe:/a:redhat:ai_inference_server:3","status":"affected","packages":["rhaii/vllm-gaudi-rhel9","rhaii/vllm-neuron-rhel9","rhaii/vllm-tpu-rhel9","rhaiis/vllm-cpu-rhel9","rhaiis/vllm-cuda-rhel9","rhaiis/vllm-neuron-rhel9","rhaiis/vllm-rocm-rhel9","rhaiis/vllm-spyre-rhel9","rhaiis/vllm-tpu-rhel9"],"package_count":9,"fixed":[],"advisories":[],"note":"Will not fix","provider":"redhat-vex","source":"CVE-2026-94627"},{"vendor":"Red Hat","product":"Red Hat Enterprise Linux AI (RHEL AI) 3","cpe":"cpe:/a:redhat:enterprise_linux_ai:3","status":"affected","packages":["rhelai3/bootc-aws-cuda-rhel9","rhelai3/bootc-azure-cuda-rhel9","rhelai3/bootc-azure-rocm-rhel9","rhelai3/bootc-cuda-rhel9","rhelai3/bootc-gcp-cuda-rhel9","rhelai3/bootc-rocm-rhel9"],"package_count":6,"fixed":[],"advisories":[],"note":"Affected","provider":"redhat-vex","source":"CVE-2026-94627"},{"vendor":"Red Hat","product":"Red Hat Enterprise Linux AI (RHEL AI) 3","cpe":"cpe:/a:redhat:enterprise_linux_ai:3","status":"affected","packages":["rhelai3/bootc-gaudi-rhel9"],"package_count":1,"fixed":[],"advisories":[],"note":"Will not fix","provider":"redhat-vex","source":"CVE-2026-94627"},{"vendor":"Red Hat","product":"Red Hat OpenShift AI (RHOAI)","cpe":"cpe:/a:redhat:openshift_ai","status":"affected","packages":["rhoai/odh-llm-d-kv-cache-rhel9"],"package_count":1,"fixed":[],"advisories":[],"note":"Affected","provider":"redhat-vex","source":"CVE-2026-94627"},{"vendor":"Red Hat","product":"Red Hat OpenShift AI (RHOAI)","cpe":"cpe:/a:redhat:openshift_ai","status":"affected","packages":["rhoai/odh-vllm-gaudi-rhel9"],"package_count":1,"fixed":[],"advisories":[],"note":"Will not fix","provider":"redhat-vex","source":"CVE-2026-94627"},{"vendor":"Red Hat","product":"Red Hat OpenShift AI (RHOAI)","cpe":"cpe:/a:redhat:openshift_ai","status":"not_affected","packages":["rhoai/odh-th-torch-cuda-py312-rhel9"],"package_count":1,"fixed":[],"advisories":[],"note":"vulnerable_code_not_in_execute_path","provider":"redhat-vex","source":"CVE-2026-94627"},{"vendor":"Red Hat","product":"Red Hat OpenShift AI 3.4","cpe":"cpe:/a:redhat:openshift_ai:3.4::el9","status":"fixed","packages":["rhoai/odh-kserve-agent-rhel9","rhoai/odh-kserve-controller-rhel9","rhoai/odh-kserve-router-rhel9","rhoai/odh-kserve-storage-initializer-rhel9"],"package_count":4,"fixed":[],"advisories":["RHSA-2026:84734"],"provider":"redhat-vex","source":"CVE-2026-94627"},{"vendor":"Red Hat","product":"Red Hat OpenShift AI 3.4","cpe":"cpe:/a:redhat:openshift_ai:3.4::el9","status":"not_affected","packages":["rhoai/odh-ai-gateway-payload-processing-rhel9","rhoai/odh-automl-rhel9","rhoai/odh-autorag-rhel9","rhoai/odh-built-in-detector-rhel9","rhoai/odh-cli-rhel9","rhoai/odh-dashboard-rhel9","rhoai/odh-data-science-pipelines-argo-argoexec-rhel9","rhoai/odh-data-science-pipelines-argo-workflowcontroller-rhel9","rhoai/odh-data-science-pipelines-operator-controller-rhel9","rhoai/odh-eval-hub-rhel9","rhoai/odh-feast-operator-rhel9","rhoai/odh-feature-server-rhel9","rhoai/odh-fms-guardrails-orchestrator-rhel9","rhoai/odh-guardrails-detector-huggingface-runtime-rhel9","rhoai/odh-kf-notebook-controller-rhel9","rhoai/odh-kserve-autogluon-server-rhel9","rhoai/odh-kserve-llmisvc-controller-rhel9","rhoai/odh-kube-auth-proxy-rhel9","rhoai/odh-kuberay-operator-controller-rhel9","rhoai/odh-llama-stack-core-rhel9","rhoai/odh-llama-stack-k8s-operator-rhel9","rhoai/odh-llm-d-batch-gateway-apiserver-rhel9","rhoai/odh-llm-d-batch-gateway-gc-rhel9","rhoai/odh-llm-d-batch-gateway-processor-rhel9","rhoai/odh-llm-d-inference-scheduler-rhel9"],"package_count":102,"fixed":[],"advisories":[],"note":"vulnerable_code_not_present","provider":"redhat-vex","source":"CVE-2026-94627"}],"references":[{"type":"WEB","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-94627.json"},{"type":"WEB","url":"https://access.redhat.com/security/cve/CVE-2026-94627"},{"type":"WEB","url":"https://bugzilla.redhat.com/show_bug.cgi?id=2537919"},{"type":"WEB","url":"https://www.cve.org/CVERecord?id=CVE-2026-94627"},{"type":"WEB","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-94627"},{"type":"WEB","url":"https://github.com/vllm-project/vllm"},{"type":"WEB","url":"https://github.com/vllm-project/vllm/blob/v0.29.0/vllm/distributed/kv_transfer/kv_connector/v1/mooncake/mooncake_connector.py#L1978-L1989"},{"type":"WEB","url":"https://github.com/vllm-project/vllm/pull/49796"},{"type":"WEB","url":"https://www.vulncheck.com/advisories/vllm-through-0.29.0-gpu-kv-cache-leak-via-mooncake-transfer-id-collision"}]}],"references":[{"type":"FIX","url":"https://github.com/vllm-project/vllm/pull/49796"},{"type":"WEB","url":"https://github.com/vllm-project/vllm/blob/v0.29.0/vllm/distributed/kv_transfer/kv_connector/v1/mooncake/mooncake_connector.py#L1978-L1989"},{"type":"WEB","url":"https://github.com/vllm-project/vllm"},{"type":"WEB","url":"https://www.vulncheck.com/advisories/vllm-through-0.29.0-gpu-kv-cache-leak-via-mooncake-transfer-id-collision"},{"type":"WEB","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-94627.json"},{"type":"WEB","url":"https://access.redhat.com/security/cve/CVE-2026-94627"},{"type":"WEB","url":"https://bugzilla.redhat.com/show_bug.cgi?id=2537919"},{"type":"WEB","url":"https://www.cve.org/CVERecord?id=CVE-2026-94627"},{"type":"WEB","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-94627"}],"sources":[{"provider":"csaf-bsi","source_id":"WID-SEC-W-2026-3507","url":"https://wid.cert-bund.de/portal/wid/securityadvisory?name=WID-SEC-2026-3507","license":"TLP:CLEAR, mirroring permitted","modified_at":"2026-11-09T23:00:00Z","relation":"upstream"},{"provider":"cvelistv5","source_id":"CVE-2026-94627","url":"https://www.cve.org/CVERecord?id=CVE-2026-94627","license":"CC-BY-4.0","modified_at":"2026-09-22T12:51:23.062Z","relation":"self"},{"provider":"nvd","source_id":"CVE-2026-94627","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-94627","relation":"self"},{"provider":"redhat-vex","source_id":"CVE-2026-94627","url":"https://access.redhat.com/security/cve/CVE-2026-94627","license":"CC-BY-4.0","modified_at":"2026-09-25T15:51:42Z","relation":"self"}]}
