{"schema":"vulnpedia.research/1","cve":"CVE-2026-100649","generated":"2026-10-26T13:52:32.101Z","references":[{"url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-j682-9xp5-rrf3","tags":["vendor-advisory","Vendor Advisory","vendor-reference"],"source":"mitre","sourceId":"VulnCheck","modified":"2026-09-26T13:23:19.687Z","origins":[{"source":"mitre","sourceId":"VulnCheck","modified":"2026-09-26T13:23:19.687Z","url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-j682-9xp5-rrf3"},{"source":"nvd","sourceId":"CVE-2026-100649","modified":"2026-10-06T19:37:35.050","url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-j682-9xp5-rrf3"},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-j682-9xp5-rrf3"}]},{"url":"https://www.vulncheck.com/advisories/vllm-before-0.29.0-resource-limit-bypass-via-sampler-subclass","tags":["third-party-advisory","Third Party Advisory","vendor-reference"],"source":"mitre","sourceId":"VulnCheck","modified":"2026-09-26T13:23:19.687Z","origins":[{"source":"mitre","sourceId":"VulnCheck","modified":"2026-09-26T13:23:19.687Z","url":"https://www.vulncheck.com/advisories/vllm-before-0.29.0-resource-limit-bypass-via-sampler-subclass"},{"source":"nvd","sourceId":"CVE-2026-100649","modified":"2026-10-06T19:37:35.050","url":"https://www.vulncheck.com/advisories/vllm-before-0.29.0-resource-limit-bypass-via-sampler-subclass"},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://www.vulncheck.com/advisories/vllm-before-0.29.0-resource-limit-bypass-via-sampler-subclass"}]},{"url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","tags":["vendor-advisory"],"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"url":"https://access.redhat.com/security/cve/CVE-2026-100649","tags":["vendor-reference"],"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://access.redhat.com/security/cve/CVE-2026-100649"}]},{"url":"https://bugzilla.redhat.com/show_bug.cgi?id=2541914","tags":["vendor-reference"],"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://bugzilla.redhat.com/show_bug.cgi?id=2541914"}]},{"url":"https://www.cve.org/CVERecord?id=CVE-2026-100649","tags":["vendor-reference"],"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://www.cve.org/CVERecord?id=CVE-2026-100649"}]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-100649","tags":["vendor-reference"],"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-100649"}]},{"source":"euvd","sourceId":"EUVD-2026-87743","url":"https://euvd.enisa.europa.eu/vulnerability/EUVD-2026-87743","tags":["ADVISORY"],"origins":[{"source":"euvd","sourceId":"EUVD-2026-87743","url":"https://euvd.enisa.europa.eu/vulnerability/EUVD-2026-87743"}]}],"packages":[],"pocCandidates":[],"cataloguedPoc":[],"vendorStatements":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_ai_inference_server:rhaii/vllm-cpu-rhel9","product":"rhaii/vllm-cpu-rhel9 as a component of Red Hat AI Inference Server","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/vllm-cpu-rhel9@sha256:81c5994dc20321f8bd3969099eade78ef127cbd79f2d0ccea101c773e4c7fa4d?repository_url=registry.redhat.io/rhaii/vllm-cpu-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_ai_inference_server:rhaii/vllm-cuda-rhel9","product":"rhaii/vllm-cuda-rhel9 as a component of Red Hat AI Inference Server","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/vllm-cuda-rhel9@sha256:0d22e2301ce26ef68effda5d3a4a20381ad0ea34d1240e159e245bd44c2e125b?repository_url=registry.redhat.io/rhaii/vllm-cuda-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_ai_inference_server:rhaii/vllm-gaudi-rhel9","product":"rhaii/vllm-gaudi-rhel9 as a component of Red Hat AI Inference Server","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/vllm-gaudi-rhel9@sha256:ce1e4fd82dd37d299ce8c64a237971d00faf602f06868c986e81aefa04c283ef?repository_url=registry.redhat.io/rhaii/vllm-gaudi-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_ai_inference_server:rhaii/vllm-neuron-rhel9","product":"rhaii/vllm-neuron-rhel9 as a component of Red Hat AI Inference Server","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/vllm-neuron-rhel9@sha256:1079c9afecdbc33f4790792f0d712dbf7523790a92dcfbe8fb5cf4c2d282fc5b?repository_url=registry.redhat.io/rhaii/vllm-neuron-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_ai_inference_server:rhaii/vllm-rocm-rhel9","product":"rhaii/vllm-rocm-rhel9 as a component of Red Hat AI Inference Server","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/vllm-rocm-rhel9@sha256:1a30359d6ca5d0891e403b9b0a23c0ae1ad6e30a2429b9b42e246b874fe8859a?repository_url=registry.redhat.io/rhaii/vllm-rocm-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_ai_inference_server:rhaii/vllm-spyre-rhel9","product":"rhaii/vllm-spyre-rhel9 as a component of Red Hat AI Inference Server","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/vllm-spyre-rhel9@sha256:11ad6e13c41728944d933c7abe01d1d10ecb3f8e8b4f0da50ec53b7df6f036a8?repository_url=registry.redhat.io/rhaii/vllm-spyre-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_ai_inference_server:rhaii/vllm-tpu-rhel9","product":"rhaii/vllm-tpu-rhel9 as a component of Red Hat AI Inference Server","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/vllm-tpu-rhel9@sha256:0c2ac747d67dc0e80162ff05867dd702e474b2217659f19f28b8e3a0c03a942a?repository_url=registry.redhat.io/rhaii/vllm-tpu-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_ai_inference_server:rhaiis/vllm-cpu-rhel9","product":"rhaiis/vllm-cpu-rhel9 as a component of Red Hat AI Inference Server","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/vllm-cpu-rhel9@sha256:fb7791bbfea99e9f81529fe8cc1e7c834ca470585c27590e3d0a6834167cd436?repository_url=registry.redhat.io/rhaiis/vllm-cpu-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_ai_inference_server:rhaiis/vllm-cuda-rhel9","product":"rhaiis/vllm-cuda-rhel9 as a component of Red Hat AI Inference Server","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/vllm-cuda-rhel9@sha256:73bbc4560132c4aaa9bbd608ce9560eeae7b30bcc8f014da460afa5622139372?repository_url=registry.redhat.io/rhaiis/vllm-cuda-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_ai_inference_server:rhaiis/vllm-neuron-rhel9","product":"rhaiis/vllm-neuron-rhel9 as a component of Red Hat AI Inference Server","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/vllm-neuron-rhel9@sha256:8ba72cec5d79b1a751eb09774c81fd6181b43ec69a2e5116982661b34366874d?repository_url=registry.redhat.io/rhaiis/vllm-neuron-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_ai_inference_server:rhaiis/vllm-rocm-rhel9","product":"rhaiis/vllm-rocm-rhel9 as a component of Red Hat AI Inference Server","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/vllm-rocm-rhel9@sha256:37b64a8df90756aae0ee96d2f07765adbce5ed5beb2414ac3845806ad2219cfc?repository_url=registry.redhat.io/rhaiis/vllm-rocm-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_ai_inference_server:rhaiis/vllm-spyre-rhel9","product":"rhaiis/vllm-spyre-rhel9 as a component of Red Hat AI Inference Server","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/vllm-spyre-rhel9@sha256:3dd7eff1c62cd4609b4c76590b77bb801cd064e40642ed32b5aa900a6bc643ea?repository_url=registry.redhat.io/rhaiis/vllm-spyre-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_ai_inference_server:rhaiis/vllm-tpu-rhel9","product":"rhaiis/vllm-tpu-rhel9 as a component of Red Hat AI Inference Server","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/vllm-tpu-rhel9@sha256:b1c12112a1e8a298645e10d34d44c9961a6121aa8d87fc5d65ad83e1ffa3c586?repository_url=registry.redhat.io/rhaiis/vllm-tpu-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_enterprise_linux_ai_(rhel_ai)_3:rhelai3/bootc-aws-cuda-rhel9","product":"rhelai3/bootc-aws-cuda-rhel9 as a component of Red Hat Enterprise Linux AI (RHEL AI) 3","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/bootc-aws-cuda-rhel9@sha256:a754d4d1126e98f203249414b3831182bb15ee6b3643452f167849defc642e4b?repository_url=registry.redhat.io/rhelai3/bootc-aws-cuda-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_enterprise_linux_ai_(rhel_ai)_3:rhelai3/bootc-azure-cuda-rhel9","product":"rhelai3/bootc-azure-cuda-rhel9 as a component of Red Hat Enterprise Linux AI (RHEL AI) 3","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/bootc-azure-cuda-rhel9@sha256:df5b6edd1bdb055219a7385b6d5923f3262c3a408f92da734e3027af55fd593f?repository_url=registry.redhat.io/rhelai3/bootc-azure-cuda-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_enterprise_linux_ai_(rhel_ai)_3:rhelai3/bootc-azure-rocm-rhel9","product":"rhelai3/bootc-azure-rocm-rhel9 as a component of Red Hat Enterprise Linux AI (RHEL AI) 3","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/bootc-azure-rocm-rhel9@sha256:8ea1233b1fb6b82d0aa69e9e4c47d2df242001280c11a61584066c27c5bec3a5?repository_url=registry.redhat.io/rhelai3/bootc-azure-rocm-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_enterprise_linux_ai_(rhel_ai)_3:rhelai3/bootc-cuda-rhel9","product":"rhelai3/bootc-cuda-rhel9 as a component of Red Hat Enterprise Linux AI (RHEL AI) 3","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/bootc-cuda-rhel9@sha256:b45f3bc25a7d3c9f3bfdc1de3be6db59f6c3109f62326d74fae14e832a797981?repository_url=registry.redhat.io/rhelai3/bootc-cuda-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_enterprise_linux_ai_(rhel_ai)_3:rhelai3/bootc-gaudi-rhel9","product":"rhelai3/bootc-gaudi-rhel9 as a component of Red Hat Enterprise Linux AI (RHEL AI) 3","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/bootc-gaudi-rhel9@sha256:e1057b59ca1be9501332abf97230855e85c7df858356be0da9dce230ea7e98e1?repository_url=registry.redhat.io/rhelai3/bootc-gaudi-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_enterprise_linux_ai_(rhel_ai)_3:rhelai3/bootc-gcp-cuda-rhel9","product":"rhelai3/bootc-gcp-cuda-rhel9 as a component of Red Hat Enterprise Linux AI (RHEL AI) 3","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/bootc-gcp-cuda-rhel9@sha256:bac5d435d156c3ddcb581d3ff05fdce9f72bd617c5b8435cede16bb17b7bf12e?repository_url=registry.redhat.io/rhelai3/bootc-gcp-cuda-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_enterprise_linux_ai_(rhel_ai)_3:rhelai3/bootc-rocm-rhel9","product":"rhelai3/bootc-rocm-rhel9 as a component of Red Hat Enterprise Linux AI (RHEL AI) 3","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/bootc-rocm-rhel9@sha256:38ce057f97034bbb8eaea574d41cc7eebaeaa9265a4cdd966476dd3fac129c3c?repository_url=registry.redhat.io/rhelai3/bootc-rocm-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_openshift_ai_(rhoai):rhoai/odh-kserve-agent-rhel9","product":"rhoai/odh-kserve-agent-rhel9 as a component of Red Hat OpenShift AI (RHOAI)","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/odh-kserve-agent-rhel9@sha256:eddf436baf8906d08f1cd28fecdc73f70f9bbf11a73e327d9c97495ee921c523?repository_url=registry.redhat.io/rhoai/odh-kserve-agent-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_openshift_ai_(rhoai):rhoai/odh-kserve-controller-rhel9","product":"rhoai/odh-kserve-controller-rhel9 as a component of Red Hat OpenShift AI (RHOAI)","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/odh-kserve-controller-rhel9@sha256:8b4d4c6a45b2ff1b25fda45dc022066ed5d055267ce08ab0b31e58191970632e?repository_url=registry.redhat.io/rhoai/odh-kserve-controller-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_openshift_ai_(rhoai):rhoai/odh-kserve-router-rhel9","product":"rhoai/odh-kserve-router-rhel9 as a component of Red Hat OpenShift AI (RHOAI)","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/odh-kserve-router-rhel9@sha256:1fb5c0b39c43904a5843a9e01cddac40cfe5580d9bb7abdf6083c9f53d1655b3?repository_url=registry.redhat.io/rhoai/odh-kserve-router-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_openshift_ai_(rhoai):rhoai/odh-kserve-storage-initializer-rhel9","product":"rhoai/odh-kserve-storage-initializer-rhel9 as a component of Red Hat OpenShift AI (RHOAI)","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/odh-kserve-storage-initializer-rhel9@sha256:7c88948bcfa0fc3439b1eae31c426639346044505aff440238bfb194fd15ce99?repository_url=registry.redhat.io/rhoai/odh-kserve-storage-initializer-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_openshift_ai_(rhoai):rhoai/odh-llm-d-kv-cache-rhel9","product":"rhoai/odh-llm-d-kv-cache-rhel9 as a component of Red Hat OpenShift AI (RHOAI)","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/odh-llm-d-kv-cache-rhel9@sha256:cb5eedeefd2578cd9bd8637250bbe299c84539f57b4ead75d630ed9f833a6ef2?repository_url=registry.redhat.io/rhoai/odh-llm-d-kv-cache-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_openshift_ai_(rhoai):rhoai/odh-th-torch-cuda-py312-rhel9","product":"rhoai/odh-th-torch-cuda-py312-rhel9 as a component of Red Hat OpenShift AI (RHOAI)","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/odh-th-torch-cuda-py312-rhel9@sha256:efd7b3607b699092c68359d20ef0d0043114191a097637667c0d2cf78fbea702?repository_url=registry.redhat.io/rhoai/odh-th-torch-cuda-py312-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]},{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json","productId":"red_hat_openshift_ai_(rhoai):rhoai/odh-vllm-gaudi-rhel9","product":"rhoai/odh-vllm-gaudi-rhel9 as a component of Red Hat OpenShift AI (RHOAI)","status":"known_affected","remediations":[{"category":"workaround","details":"1. Summary of action: Restrict network reachability to the inference server and avoid deploying models with the hardware-accelerated PyNvVideoCodec video backend.\n\n2. Command examples:\nTo block direct untrusted network access to the vLLM server port (default 8000) using firewalld:\n```\nfirewall-cmd --permanent --zone=public --remove-port=8000/tcp\nfirewall-cmd --permanent --zone=trusted --add-source=<TRUSTED_IP_OR_CIDR>\nfirewall-cmd --reload\n```\nIn OpenShift environments, deploy a NetworkPolicy to restrict ingress traffic so that only authenticated API gateways or designated client pods can reach the inference service. Additionally, avoid specifying \"backend\": \"pynvvideocodec\" in the server video configuration.\n\n3. Caveats: Restricting network access prevents direct unauthenticated requests from reaching the endpoint. Avoiding hardware-accelerated video decoding shifts media decoding workloads to software, which may increase CPU utilization and inference latency for video inputs.\n\n4. Warning: Applying firewall rule reloads takes effect immediately and may sever existing connections. Modifying deployment configurations or restarting inference pods will temporarily interrupt ongoing serving sessions."},{"category":"none_available","details":"Fix deferred"}],"purl":"pkg:oci/odh-vllm-gaudi-rhel9@sha256:87eec3862eb6ddef445f89dcf6fd89add7ef202ea7c563b1ee7347537fb3e250?repository_url=registry.redhat.io/rhoai/odh-vllm-gaudi-rhel9","origins":[{"source":"redhat","sourceId":"CVE-2026-100649","modified":"2026-10-06T23:36:00+00:00","url":"https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-100649.json"}]}],"vendorCoverage":{"total":27,"shown":27,"omitted":0,"source":"redhat"}}