--- # Source: nvsentinel/charts/labeler/templates/serviceaccount.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. apiVersion: v1 kind: ServiceAccount metadata: name: labeler labels: helm.sh/chart: labeler-0.1.0 app.kubernetes.io/name: labeler app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "1.16.0" app.kubernetes.io/managed-by: Helm --- # Source: nvsentinel/charts/metadata-collector/templates/serviceaccount.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. apiVersion: v1 kind: ServiceAccount metadata: name: metadata-collector labels: helm.sh/chart: metadata-collector-0.1.0 app.kubernetes.io/name: metadata-collector app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "1.0.0" app.kubernetes.io/managed-by: Helm --- # Source: nvsentinel/templates/serviceaccount.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. apiVersion: v1 kind: ServiceAccount metadata: name: platform-connectors labels: helm.sh/chart: nvsentinel-v1.9.0 app.kubernetes.io/name: nvsentinel app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "v1.9.0" app.kubernetes.io/managed-by: Helm --- # Source: nvsentinel/charts/gpu-health-monitor/templates/configmap.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. apiVersion: v1 kind: ConfigMap metadata: name: gpu-health-monitor labels: helm.sh/chart: gpu-health-monitor-0.1.0 app.kubernetes.io/name: gpu-health-monitor app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "1.16.0" app.kubernetes.io/managed-by: Helm data: config.ini: | [logging] LogFormat = %%(asctime)s,%%(msecs)03d %%(levelname)-8s [%%(filename)s:%%(lineno)d] %%(message)s DateTimeFormat = %%Y-%%m-%%d:%%H:%%M:%%S [dcgm] PollIntervalSeconds = 15 [eventprocessors.kubernetes] ConflictRetryCount = 10 [eventprocessors.platformconnector] SocketPath = /var/run/nvsentinel.sock [cli] EnabledEventProcessors = PlatformConnectorEventProcessor dcgmerrorsmapping.csv: |- DCGM_FR_UNKNOWN,CONTACT_SUPPORT DCGM_FR_UNRECOGNIZED,CONTACT_SUPPORT DCGM_FR_PCI_REPLAY_RATE,CONTACT_SUPPORT DCGM_FR_VOLATILE_DBE_DETECTED,COMPONENT_RESET DCGM_FR_VOLATILE_SBE_DETECTED,NONE DCGM_FR_PENDING_PAGE_RETIREMENTS,NONE DCGM_FR_RETIRED_PAGES_LIMIT,CONTACT_SUPPORT DCGM_FR_RETIRED_PAGES_DBE_LIMIT,CONTACT_SUPPORT DCGM_FR_CORRUPT_INFOROM,COMPONENT_RESET DCGM_FR_CLOCKS_EVENT_THERMAL,CONTACT_SUPPORT DCGM_FR_CLOCK_THROTTLE_THERMAL,NONE DCGM_FR_POWER_UNREADABLE,RESTART_VM DCGM_FR_CLOCKS_EVENT_POWER,NONE DCGM_FR_CLOCK_THROTTLE_POWER,NONE DCGM_FR_NVLINK_ERROR_THRESHOLD,NONE DCGM_FR_NVLINK_DOWN,RESTART_VM DCGM_FR_NVSWITCH_FATAL_ERROR,CONTACT_SUPPORT DCGM_FR_NVSWITCH_NON_FATAL_ERROR,NONE DCGM_FR_NVSWITCH_DOWN,COMPONENT_RESET DCGM_FR_NO_ACCESS_TO_FILE,CONTACT_SUPPORT DCGM_FR_NVML_API,CONTACT_SUPPORT DCGM_FR_DEVICE_COUNT_MISMATCH,CONTACT_SUPPORT DCGM_FR_BAD_PARAMETER,CONTACT_SUPPORT DCGM_FR_CANNOT_OPEN_LIB,CONTACT_SUPPORT DCGM_FR_DENYLISTED_DRIVER,CONTACT_SUPPORT DCGM_FR_NVML_LIB_BAD,CONTACT_SUPPORT DCGM_FR_GRAPHICS_PROCESSES,CONTACT_SUPPORT DCGM_FR_HOSTENGINE_CONN,CONTACT_SUPPORT DCGM_FR_FIELD_QUERY,RESTART_VM DCGM_FR_BAD_CUDA_ENV,CONTACT_SUPPORT DCGM_FR_PERSISTENCE_MODE,CONTACT_SUPPORT DCGM_FR_LOW_BANDWIDTH,CONTACT_SUPPORT DCGM_FR_HIGH_LATENCY,CONTACT_SUPPORT DCGM_FR_CANNOT_GET_FIELD_TAG,CONTACT_SUPPORT DCGM_FR_FIELD_VIOLATION,RESTART_VM DCGM_FR_FIELD_THRESHOLD,RESTART_VM DCGM_FR_FIELD_VIOLATION_DBL,RESTART_VM DCGM_FR_FIELD_THRESHOLD_DBL,RESTART_VM DCGM_FR_UNSUPPORTED_FIELD_TYPE,RESTART_VM DCGM_FR_FIELD_THRESHOLD_TS,RESTART_VM DCGM_FR_FIELD_THRESHOLD_TS_DBL,RESTART_VM DCGM_FR_THERMAL_VIOLATIONS,CONTACT_SUPPORT DCGM_FR_THERMAL_VIOLATIONS_TS,CONTACT_SUPPORT DCGM_FR_TEMP_VIOLATION,CONTACT_SUPPORT DCGM_FR_CLOCKS_EVENT_VIOLATION,RESTART_VM DCGM_FR_THROTTLING_VIOLATION,RESTART_VM DCGM_FR_INTERNAL,CONTACT_SUPPORT DCGM_FR_PCIE_GENERATION,CONTACT_SUPPORT DCGM_FR_PCIE_WIDTH,CONTACT_SUPPORT DCGM_FR_ABORTED,NONE DCGM_FR_TEST_DISABLED,NONE DCGM_FR_CANNOT_GET_STAT,CONTACT_SUPPORT DCGM_FR_STRESS_LEVEL,RESTART_VM DCGM_FR_CUDA_API,CONTACT_SUPPORT DCGM_FR_FAULTY_MEMORY,CONTACT_SUPPORT DCGM_FR_CANNOT_SET_WATCHES,RESTART_VM DCGM_FR_CUDA_UNBOUND,RESTART_VM DCGM_FR_ECC_DISABLED,CONTACT_SUPPORT DCGM_FR_MEMORY_ALLOC,RESTART_VM DCGM_FR_CUDA_DBE,CONTACT_SUPPORT DCGM_FR_MEMORY_MISMATCH,CONTACT_SUPPORT DCGM_FR_CUDA_DEVICE,CONTACT_SUPPORT DCGM_FR_ECC_UNSUPPORTED,CONTACT_SUPPORT DCGM_FR_ECC_PENDING,RESTART_VM DCGM_FR_MEMORY_BANDWIDTH,RESTART_VM DCGM_FR_TARGET_POWER,NONE DCGM_FR_API_FAIL,RESTART_VM DCGM_FR_API_FAIL_GPU,RESTART_VM DCGM_FR_CUDA_CONTEXT,CONTACT_SUPPORT DCGM_FR_DCGM_API,CONTACT_SUPPORT DCGM_FR_CONCURRENT_GPUS,CONTACT_SUPPORT DCGM_FR_TOO_MANY_ERRORS,CONTACT_SUPPORT DCGM_FR_NVLINK_CRC_ERROR_THRESHOLD,CONTACT_SUPPORT DCGM_FR_NVLINK_ERROR_CRITICAL,CONTACT_SUPPORT DCGM_FR_ENFORCED_POWER_LIMIT,CONTACT_SUPPORT DCGM_FR_MEMORY_ALLOC_HOST,RESTART_VM DCGM_FR_GPU_OP_MODE,CONTACT_SUPPORT DCGM_FR_NO_MEMORY_CLOCKS,CONTACT_SUPPORT DCGM_FR_NO_GRAPHICS_CLOCKS,NONE DCGM_FR_HAD_TO_RESTORE_STATE,RESTART_VM DCGM_FR_L1TAG_UNSUPPORTED,CONTACT_SUPPORT DCGM_FR_L1TAG_MISCOMPARE,CONTACT_SUPPORT DCGM_FR_ROW_REMAP_FAILURE,CONTACT_SUPPORT DCGM_FR_UNCONTAINED_ERROR,RESTART_VM DCGM_FR_EMPTY_GPU_LIST,CONTACT_SUPPORT DCGM_FR_DBE_PENDING_PAGE_RETIREMENTS,RESTART_VM DCGM_FR_UNCORRECTABLE_ROW_REMAP,NONE DCGM_FR_PENDING_ROW_REMAP,COMPONENT_RESET DCGM_FR_BROKEN_P2P_MEMORY_DEVICE,CONTACT_SUPPORT DCGM_FR_BROKEN_P2P_WRITER_DEVICE,CONTACT_SUPPORT DCGM_FR_NVSWITCH_NVLINK_DOWN,CONTACT_SUPPORT DCGM_FR_EUD_BINARY_PERMISSIONS,CONTACT_SUPPORT DCGM_FR_EUD_NON_ROOT_USER,CONTACT_SUPPORT DCGM_FR_EUD_SPAWN_FAILURE,CONTACT_SUPPORT DCGM_FR_EUD_TIMEOUT,CONTACT_SUPPORT DCGM_FR_EUD_ZOMBIE,CONTACT_SUPPORT DCGM_FR_EUD_NON_ZERO_EXIT_CODE,CONTACT_SUPPORT DCGM_FR_EUD_TEST_FAILED,CONTACT_SUPPORT DCGM_FR_FILE_CREATE_PERMISSIONS,CONTACT_SUPPORT DCGM_FR_PAUSE_RESUME_FAILED,CONTACT_SUPPORT DCGM_FR_PCIE_H_REPLAY_VIOLATION,CONTACT_SUPPORT DCGM_FR_GPU_EXPECTED_NVLINKS_UP,CONTACT_SUPPORT DCGM_FR_NVSWITCH_EXPECTED_NVLINKS_UP,CONTACT_SUPPORT DCGM_FR_XID_ERROR,NONE DCGM_FR_SBE_VIOLATION,CONTACT_SUPPORT DCGM_FR_DBE_VIOLATION,CONTACT_SUPPORT DCGM_FR_PCIE_REPLAY_VIOLATION,CONTACT_SUPPORT DCGM_FR_SBE_THRESHOLD_VIOLATION,CONTACT_SUPPORT DCGM_FR_DBE_THRESHOLD_VIOLATION,CONTACT_SUPPORT DCGM_FR_PCIE_REPLAY_THRESHOLD_VIOLATION,CONTACT_SUPPORT DCGM_FR_CUDA_FM_NOT_INITIALIZED,CONTACT_SUPPORT DCGM_FR_SXID_ERROR,RESTART_VM DCGM_FR_GFLOPS_THRESHOLD_VIOLATION,CONTACT_SUPPORT DCGM_FR_NAN_VALUE,CONTACT_SUPPORT DCGM_FR_FABRIC_MANAGER_TRAINING_ERROR,CONTACT_SUPPORT DCGM_FR_BROKEN_P2P_PCIE_MEMORY_DEVICE,CONTACT_SUPPORT DCGM_FR_BROKEN_P2P_PCIE_WRITER_DEVICE,CONTACT_SUPPORT DCGM_FR_BROKEN_P2P_NVLINK_MEMORY_DEVICE,CONTACT_SUPPORT DCGM_FR_BROKEN_P2P_NVLINK_WRITER_DEVICE,CONTACT_SUPPORT DCGM_FR_TEST_SKIPPED,NONE DCGM_FR_SRAM_THRESHOLD,CONTACT_SUPPORT DCGM_FR_NVLINK_EFFECTIVE_BER_THRESHOLD,CONTACT_SUPPORT DCGM_FR_FALLEN_OFF_BUS,RESTART_BM DCGM_FR_NVLINK_SYMBOL_BER_THRESHOLD,CONTACT_SUPPORT DCGM_FR_IMEX_UNHEALTHY,CONTACT_SUPPORT DCGM_FR_FABRIC_PROBE_STATE,CONTACT_SUPPORT DCGM_FR_BINARY_PERMISSIONS,CONTACT_SUPPORT DCGM_FR_GPU_RECOVERY_RESET,COMPONENT_RESET DCGM_FR_GPU_RECOVERY_REBOOT,RESTART_BM DCGM_FR_GPU_RECOVERY_DRAIN_P2P,CONTACT_SUPPORT DCGM_FR_GPU_RECOVERY_DRAIN_RESET,COMPONENT_RESET DCGM_FR_NCCL_ERROR,COMPONENT_RESET DCGM_FR_ERROR_SENTINEL,NONE --- # Source: nvsentinel/charts/syslog-health-monitor/templates/configmap.yaml # Copyright (c) 2026, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # Single ConfigMap holding every file configuration owned by the # syslog-health-monitor pod. Each data key becomes a file under the mount # directory (/etc/syslog-health-monitor/) and is referenced by the matching # CLI flag in the DaemonSet args. apiVersion: v1 kind: ConfigMap metadata: name: 'syslog-health-monitor-config' labels: helm.sh/chart: syslog-health-monitor-0.1.0 app.kubernetes.io/name: syslog-health-monitor app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "1.16.0" app.kubernetes.io/managed-by: Helm data: nic-driver.toml: | [[nicDriverDetection.patterns]] name = "cmd_exec_timeout" enabled = false [[nicDriverDetection.patterns]] name = "health_poll_failed" enabled = false [[nicDriverDetection.patterns]] name = "unrecoverable_err" enabled = false [[nicDriverDetection.patterns]] name = "netdev_watchdog" enabled = false [[nicDriverDetection.patterns]] name = "pci_power_insufficient" enabled = false [[nicDriverDetection.patterns]] name = "port_module_high_temp" enabled = false [[nicDriverDetection.patterns]] name = "access_reg_failed" enabled = false [[nicDriverDetection.patterns]] name = "module_unplugged" enabled = false cancellations.toml: | [[checks]] name = "SysLogsXIDError" enabled = true [[checks.cancellations]] onErrorCode = "162" cancelErrorCodes = ["163"] --- # Source: nvsentinel/templates/configmap.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. apiVersion: v1 kind: ConfigMap metadata: name: platform-connectors labels: helm.sh/chart: nvsentinel-v1.9.0 app.kubernetes.io/name: nvsentinel app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "v1.9.0" app.kubernetes.io/managed-by: Helm data: config.json: | { "enableK8sPlatformConnector": "true", "K8sConnectorQps": 5.00, "K8sConnectorBurst": 10, "MaxNodeConditionMessageLength": 1024, "CompactedHealthEventMsgLen": 72, "StoreConnectorMaxRetries": 3, "enableMongoDBStorePlatformConnector": "", "enablePostgresDBStorePlatformConnector": "false" ,"enableGRPCSinkConnector": "false" ,"GRPCSinkTarget": "" ,"GRPCSinkConnectorMaxRetries": 3 ,"GRPCSinkTokenPath": "" ,"pipeline": [{"config":"/etc/config/metadata.toml","enabled":false,"name":"MetadataAugmentor"},{"config":"/etc/config/overrides.toml","enabled":false,"name":"OverrideTransformer"}] } metadata.toml: | cacheSize = 50 cacheTTL = "3600s" allowedLabels = ["topology.kubernetes.io/zone","topology.kubernetes.io/region","node.kubernetes.io/instance-type","nvidia.com/cuda.driver-version.major","nvidia.com/cuda.driver-version.minor","nvidia.com/cuda.driver-version.revision","nvidia.com/cuda.driver-version.full","nvidia.com/cuda.runtime-version.major","nvidia.com/cuda.runtime-version.minor","nvidia.com/cuda.runtime-version.full","topology.k8s.aws/capacity-block-id","topology.k8s.aws/network-node-layer-1","topology.k8s.aws/network-node-layer-2","topology.k8s.aws/network-node-layer-3","oci.oraclecloud.com/host.id","oci.oraclecloud.com/host.network_block_id","oci.oraclecloud.com/host.rack_id","oci.oraclecloud.com/host.serial_number","cloud.google.com/gce-topology-block","cloud.google.com/gce-topology-host","cloud.google.com/gce-topology-subblock","network.topology.nvidia.com/accelerator","network.topology.nvidia.com/leaf","network.topology.nvidia.com/spine","network.topology.nvidia.com/core"] overrides.toml: | enabled = false --- # Source: nvsentinel/charts/labeler/templates/clusterrole.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. apiVersion: rbac.authorization.k8s.io/v1 kind: ClusterRole metadata: name: labeler labels: helm.sh/chart: labeler-0.1.0 app.kubernetes.io/name: labeler app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "1.16.0" app.kubernetes.io/managed-by: Helm rules: - apiGroups: - "" resources: - pods verbs: - get - list - watch - apiGroups: - "" resources: - nodes verbs: - get - list - watch - patch - update --- # Source: nvsentinel/charts/metadata-collector/templates/clusterrole.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. apiVersion: rbac.authorization.k8s.io/v1 kind: ClusterRole metadata: name: metadata-collector labels: helm.sh/chart: metadata-collector-0.1.0 app.kubernetes.io/name: metadata-collector app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "1.0.0" app.kubernetes.io/managed-by: Helm rules: - apiGroups: - "" resources: - pods verbs: - patch - apiGroups: - "" resources: - nodes/proxy verbs: - get --- # Source: nvsentinel/templates/clusterrole.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. apiVersion: rbac.authorization.k8s.io/v1 kind: ClusterRole metadata: name: platform-connectors labels: helm.sh/chart: nvsentinel-v1.9.0 app.kubernetes.io/name: nvsentinel app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "v1.9.0" app.kubernetes.io/managed-by: Helm rules: - apiGroups: - "" resources: - nodes verbs: - get - list - apiGroups: - "" resources: - nodes/status verbs: - get - patch - update - list - apiGroups: - "" resources: - events verbs: - create - update - list - apiGroups: - "" resources: - pods/eviction verbs: - create - apiGroups: - "" resources: - pods verbs: - get - list - watch --- # Source: nvsentinel/charts/labeler/templates/clusterrolebinding.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. apiVersion: rbac.authorization.k8s.io/v1 kind: ClusterRoleBinding metadata: name: labeler labels: helm.sh/chart: labeler-0.1.0 app.kubernetes.io/name: labeler app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "1.16.0" app.kubernetes.io/managed-by: Helm subjects: - kind: ServiceAccount name: labeler namespace: nvsentinel roleRef: kind: ClusterRole name: labeler apiGroup: rbac.authorization.k8s.io --- # Source: nvsentinel/charts/metadata-collector/templates/clusterrolebinding.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. apiVersion: rbac.authorization.k8s.io/v1 kind: ClusterRoleBinding metadata: name: metadata-collector labels: helm.sh/chart: metadata-collector-0.1.0 app.kubernetes.io/name: metadata-collector app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "1.0.0" app.kubernetes.io/managed-by: Helm roleRef: apiGroup: rbac.authorization.k8s.io kind: ClusterRole name: metadata-collector subjects: - kind: ServiceAccount name: metadata-collector namespace: nvsentinel --- # Source: nvsentinel/templates/clusterrolebinding.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. apiVersion: rbac.authorization.k8s.io/v1 kind: ClusterRoleBinding metadata: name: platform-connectors labels: helm.sh/chart: nvsentinel-v1.9.0 app.kubernetes.io/name: nvsentinel app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "v1.9.0" app.kubernetes.io/managed-by: Helm roleRef: apiGroup: rbac.authorization.k8s.io kind: ClusterRole name: platform-connectors subjects: - kind: ServiceAccount name: platform-connectors namespace: nvsentinel --- # Source: nvsentinel/charts/gpu-health-monitor/templates/daemonset-dcgm-3.x.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. apiVersion: apps/v1 kind: DaemonSet metadata: name: gpu-health-monitor-dcgm-3.x labels: helm.sh/chart: gpu-health-monitor-0.1.0 app.kubernetes.io/name: gpu-health-monitor app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "1.16.0" app.kubernetes.io/managed-by: Helm spec: updateStrategy: type: RollingUpdate rollingUpdate: maxUnavailable: 5% selector: matchLabels: app.kubernetes.io/name: gpu-health-monitor app.kubernetes.io/instance: nvsentinel template: metadata: annotations: checksum/config: 4c0ffba1f8b9a376165a661598a494587555926b1a0c07ad4215cad8cd215b6c labels: app.kubernetes.io/name: gpu-health-monitor app.kubernetes.io/instance: nvsentinel spec: containers: - name: gpu-health-monitor command: ["gpu_health_monitor"] args: - --config-file - "/etc/dcgmhealth/config.ini" - --dcgm-error-mapping-config-file - "/etc/dcgmhealth/dcgmerrorsmapping.csv" - --dcgm-addr - "nvidia-dcgm.gpu-operator.svc:5555" - --port - "2112" - --verbose - "False" - --state-file - "/var/run/statefile" - --dcgm-k8s-service-enabled - "true" - --metadata-path - "/var/lib/nvsentinel/gpu_metadata.json" - --processing-strategy - "EXECUTE_REMEDIATION" securityContext: runAsUser: 0 image: "ghcr.io/nvidia/nvsentinel/gpu-health-monitor:v1.9.0-dcgm-3.x" imagePullPolicy: IfNotPresent resources: limits: cpu: 500m memory: 512Mi requests: cpu: 100m memory: 128Mi ports: - name: metrics containerPort: 2112 livenessProbe: httpGet: path: /metrics port: 2112 initialDelaySeconds: 30 periodSeconds: 30 timeoutSeconds: 3 failureThreshold: 3 readinessProbe: httpGet: path: /metrics port: 2112 initialDelaySeconds: 10 periodSeconds: 10 timeoutSeconds: 3 failureThreshold: 3 volumeMounts: - name: config-vol mountPath: /etc/dcgmhealth - name: var-run-vol mountPath: /var/run/ - name: gpu-metadata mountPath: /var/lib/nvsentinel readOnly: true env: - name: NODE_NAME valueFrom: fieldRef: apiVersion: v1 fieldPath: spec.nodeName volumes: - name: config-vol configMap: name: gpu-health-monitor defaultMode: 0777 - name: var-run-vol hostPath: path: /var/run/nvsentinel type: Directory - name: gpu-metadata hostPath: path: /var/lib/nvsentinel type: DirectoryOrCreate nodeSelector: nvsentinel.dgxc.nvidia.com/dcgm.version: "3.x" tolerations: - operator: Exists --- # Source: nvsentinel/charts/gpu-health-monitor/templates/daemonset-dcgm-4.x.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. apiVersion: apps/v1 kind: DaemonSet metadata: name: gpu-health-monitor-dcgm-4.x labels: helm.sh/chart: gpu-health-monitor-0.1.0 app.kubernetes.io/name: gpu-health-monitor app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "1.16.0" app.kubernetes.io/managed-by: Helm spec: updateStrategy: type: RollingUpdate rollingUpdate: maxUnavailable: 5% selector: matchLabels: app.kubernetes.io/name: gpu-health-monitor app.kubernetes.io/instance: nvsentinel template: metadata: annotations: checksum/config: 4c0ffba1f8b9a376165a661598a494587555926b1a0c07ad4215cad8cd215b6c labels: app.kubernetes.io/name: gpu-health-monitor app.kubernetes.io/instance: nvsentinel spec: containers: - name: gpu-health-monitor command: ["gpu_health_monitor"] args: - --config-file - "/etc/dcgmhealth/config.ini" - --dcgm-error-mapping-config-file - "/etc/dcgmhealth/dcgmerrorsmapping.csv" - --dcgm-addr - "nvidia-dcgm.gpu-operator.svc:5555" - --port - "2112" - --verbose - "False" - --state-file - "/var/run/statefile" - --dcgm-k8s-service-enabled - "true" - --metadata-path - "/var/lib/nvsentinel/gpu_metadata.json" - --processing-strategy - "EXECUTE_REMEDIATION" securityContext: runAsUser: 0 image: "ghcr.io/nvidia/nvsentinel/gpu-health-monitor:v1.9.0-dcgm-4.x" imagePullPolicy: IfNotPresent resources: limits: cpu: 500m memory: 512Mi requests: cpu: 100m memory: 128Mi ports: - name: metrics containerPort: 2112 livenessProbe: httpGet: path: /metrics port: 2112 initialDelaySeconds: 30 periodSeconds: 30 timeoutSeconds: 3 failureThreshold: 3 readinessProbe: httpGet: path: /metrics port: 2112 initialDelaySeconds: 10 periodSeconds: 10 timeoutSeconds: 3 failureThreshold: 3 volumeMounts: - name: config-vol mountPath: /etc/dcgmhealth - name: var-run-vol mountPath: /var/run/ - name: gpu-metadata mountPath: /var/lib/nvsentinel readOnly: true env: - name: NODE_NAME valueFrom: fieldRef: apiVersion: v1 fieldPath: spec.nodeName volumes: - name: config-vol configMap: name: gpu-health-monitor defaultMode: 0777 - name: var-run-vol hostPath: path: /var/run/nvsentinel type: Directory - name: gpu-metadata hostPath: path: /var/lib/nvsentinel type: DirectoryOrCreate nodeSelector: nvsentinel.dgxc.nvidia.com/dcgm.version: "4.x" tolerations: - operator: Exists --- # Source: nvsentinel/charts/metadata-collector/templates/daemonset.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. apiVersion: apps/v1 kind: DaemonSet metadata: name: metadata-collector labels: helm.sh/chart: metadata-collector-0.1.0 app.kubernetes.io/name: metadata-collector app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "1.0.0" app.kubernetes.io/managed-by: Helm spec: updateStrategy: type: RollingUpdate rollingUpdate: maxUnavailable: 5% selector: matchLabels: app.kubernetes.io/name: metadata-collector app.kubernetes.io/instance: nvsentinel template: metadata: labels: app.kubernetes.io/name: metadata-collector app.kubernetes.io/instance: nvsentinel spec: serviceAccountName: metadata-collector hostNetwork: true hostPID: true containers: - name: metadata-collector securityContext: runAsUser: 0 runAsGroup: 0 privileged: true image: "ghcr.io/nvidia/nvsentinel/metadata-collector:v1.9.0" imagePullPolicy: IfNotPresent args: - --output-path=/var/lib/nvsentinel/gpu_metadata.json env: - name: NODE_NAME valueFrom: fieldRef: apiVersion: v1 fieldPath: spec.nodeName - name: KUBELET_HOST valueFrom: fieldRef: fieldPath: status.hostIP resources: limits: cpu: 500m memory: 256Mi requests: cpu: 100m memory: 128Mi volumeMounts: - name: output mountPath: /var/lib/nvsentinel - name: sys mountPath: /sys readOnly: true - mountPath: /var/lib/kubelet/pod-resources name: pod-gpu-resources readOnly: true volumes: - name: output hostPath: path: /var/lib/nvsentinel type: DirectoryOrCreate - name: sys hostPath: path: /sys type: Directory - name: pod-gpu-resources hostPath: path: /var/lib/kubelet/pod-resources type: Directory nodeSelector: nvidia.com/gpu.present: "true" nvsentinel.dgxc.nvidia.com/driver.installed: "true" tolerations: - operator: Exists runtimeClassName: nvidia --- # Source: nvsentinel/charts/syslog-health-monitor/templates/daemonset-kata.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. apiVersion: apps/v1 kind: DaemonSet metadata: name: syslog-health-monitor-kata labels: helm.sh/chart: syslog-health-monitor-0.1.0 app.kubernetes.io/name: syslog-health-monitor app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "1.16.0" app.kubernetes.io/managed-by: Helm spec: updateStrategy: type: RollingUpdate rollingUpdate: maxUnavailable: 5% selector: matchLabels: app.kubernetes.io/name: syslog-health-monitor app.kubernetes.io/instance: nvsentinel nvsentinel.dgxc.nvidia.com/kata: "true" template: metadata: labels: app.kubernetes.io/name: syslog-health-monitor app.kubernetes.io/instance: nvsentinel nvsentinel.dgxc.nvidia.com/kata: "true" spec: containers: - name: syslog-health-monitor securityContext: runAsUser: 0 capabilities: add: ["SYSLOG", "SYS_ADMIN"] image: "ghcr.io/nvidia/nvsentinel/syslog-health-monitor:v1.9.0" imagePullPolicy: IfNotPresent args: - "--polling-interval" - "15s" - "--metrics-port" - "2112" - "--kata-enabled" - "true" - "--checks" - "SysLogsXIDError,SysLogsSXIDError,SysLogsGPUFallenOff" - "--metadata-path" - "/var/lib/nvsentinel/gpu_metadata.json" - "--processing-strategy" - "EXECUTE_REMEDIATION" - "--nic-driver-config" - "/etc/syslog-health-monitor/nic-driver.toml" - "--sysfs-root" - "/nvsentinel/sys" - "--cancellations-config" - "/etc/syslog-health-monitor/cancellations.toml" resources: limits: cpu: 500m memory: 512Mi requests: cpu: 100m memory: 128Mi ports: - name: metrics containerPort: 2112 livenessProbe: httpGet: path: /metrics port: 2112 initialDelaySeconds: 30 periodSeconds: 30 timeoutSeconds: 3 failureThreshold: 3 readinessProbe: httpGet: path: /metrics port: 2112 initialDelaySeconds: 10 periodSeconds: 10 timeoutSeconds: 3 failureThreshold: 3 env: - name: NODE_NAME valueFrom: fieldRef: apiVersion: v1 fieldPath: spec.nodeName - name: LOG_LEVEL value: "info" volumeMounts: - name: var-run-vol mountPath: /var/run/ - name: syslog-state-vol mountPath: /var/run/syslog_health_monitor - name: metadata-vol mountPath: /var/lib/nvsentinel readOnly: true # Kata mode: Mount systemd journal for accessing host logs - name: host-journal mountPath: /nvsentinel/var/log/journal readOnly: true - name: host-systemd mountPath: /run/systemd/journal readOnly: true - name: host-machine-id mountPath: /etc/machine-id readOnly: true - name: proc-vol mountPath: /nvsentinel/proc readOnly: true - name: sys-vol mountPath: /nvsentinel/sys readOnly: true - name: syslog-health-monitor-config mountPath: /etc/syslog-health-monitor readOnly: true volumes: - name: var-run-vol hostPath: path: /var/run/nvsentinel type: DirectoryOrCreate - name: syslog-state-vol hostPath: path: /var/run/syslog_health_monitor type: DirectoryOrCreate - name: metadata-vol hostPath: path: /var/lib/nvsentinel type: DirectoryOrCreate # Kata mode: Systemd journal volumes for host log access - name: host-journal hostPath: path: /var/log/journal type: Directory - name: host-systemd hostPath: path: /run/systemd/journal type: Directory - name: host-machine-id hostPath: path: /etc/machine-id type: File - name: sys-vol hostPath: path: /sys type: Directory - name: syslog-health-monitor-config configMap: name: syslog-health-monitor-config - name: proc-vol hostPath: path: /proc type: Directory nodeSelector: nvsentinel.dgxc.nvidia.com/driver.installed: "true" nvsentinel.dgxc.nvidia.com/kata.enabled: "true" tolerations: - operator: Exists --- # Source: nvsentinel/charts/syslog-health-monitor/templates/daemonset-regular.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. apiVersion: apps/v1 kind: DaemonSet metadata: name: syslog-health-monitor-regular labels: helm.sh/chart: syslog-health-monitor-0.1.0 app.kubernetes.io/name: syslog-health-monitor app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "1.16.0" app.kubernetes.io/managed-by: Helm spec: updateStrategy: type: RollingUpdate rollingUpdate: maxUnavailable: 5% selector: matchLabels: app.kubernetes.io/name: syslog-health-monitor app.kubernetes.io/instance: nvsentinel nvsentinel.dgxc.nvidia.com/kata: "false" template: metadata: labels: app.kubernetes.io/name: syslog-health-monitor app.kubernetes.io/instance: nvsentinel nvsentinel.dgxc.nvidia.com/kata: "false" spec: containers: - name: syslog-health-monitor securityContext: runAsUser: 0 capabilities: add: ["SYSLOG", "SYS_ADMIN"] image: "ghcr.io/nvidia/nvsentinel/syslog-health-monitor:v1.9.0" imagePullPolicy: IfNotPresent args: - "--polling-interval" - "15s" - "--metrics-port" - "2112" - "--kata-enabled" - "false" - "--checks" - "SysLogsXIDError,SysLogsSXIDError,SysLogsGPUFallenOff" - "--metadata-path" - "/var/lib/nvsentinel/gpu_metadata.json" - "--processing-strategy" - "EXECUTE_REMEDIATION" - "--nic-driver-config" - "/etc/syslog-health-monitor/nic-driver.toml" - "--sysfs-root" - "/nvsentinel/sys" - "--cancellations-config" - "/etc/syslog-health-monitor/cancellations.toml" resources: limits: cpu: 500m memory: 512Mi requests: cpu: 100m memory: 128Mi ports: - name: metrics containerPort: 2112 livenessProbe: httpGet: path: /metrics port: 2112 initialDelaySeconds: 30 periodSeconds: 30 timeoutSeconds: 3 failureThreshold: 3 readinessProbe: httpGet: path: /metrics port: 2112 initialDelaySeconds: 10 periodSeconds: 10 timeoutSeconds: 3 failureThreshold: 3 env: - name: NODE_NAME valueFrom: fieldRef: apiVersion: v1 fieldPath: spec.nodeName - name: LOG_LEVEL value: "info" volumeMounts: - name: var-run-vol mountPath: /var/run/ - name: syslog-state-vol mountPath: /var/run/syslog_health_monitor - name: metadata-vol mountPath: /var/lib/nvsentinel readOnly: true # Regular mode: Mount journal from user-defined host path - name: var-log-vol mountPath: /nvsentinel/var/log readOnly: true - name: proc-vol mountPath: /nvsentinel/proc readOnly: true - name: sys-vol mountPath: /nvsentinel/sys readOnly: true - name: syslog-health-monitor-config mountPath: /etc/syslog-health-monitor readOnly: true volumes: - name: var-run-vol hostPath: path: /var/run/nvsentinel type: DirectoryOrCreate - name: syslog-state-vol hostPath: path: /var/run/syslog_health_monitor type: DirectoryOrCreate - name: metadata-vol hostPath: path: /var/lib/nvsentinel type: DirectoryOrCreate # Regular mode: Mount journal from user-defined host path - name: var-log-vol hostPath: path: /var/log type: Directory - name: sys-vol hostPath: path: /sys type: Directory - name: syslog-health-monitor-config configMap: name: syslog-health-monitor-config - name: proc-vol hostPath: path: /proc type: Directory nodeSelector: nvsentinel.dgxc.nvidia.com/driver.installed: "true" nvsentinel.dgxc.nvidia.com/kata.enabled: "false" tolerations: - operator: Exists --- # Source: nvsentinel/templates/daemonset.yaml apiVersion: apps/v1 kind: DaemonSet metadata: name: platform-connectors labels: helm.sh/chart: nvsentinel-v1.9.0 app.kubernetes.io/name: nvsentinel app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "v1.9.0" app.kubernetes.io/managed-by: Helm spec: updateStrategy: type: RollingUpdate rollingUpdate: maxUnavailable: 5% selector: matchLabels: app.kubernetes.io/name: nvsentinel app.kubernetes.io/instance: nvsentinel template: metadata: annotations: checksum/config: f7f3986ae45378eea88679ad99ea0bc4a0c6377c97c7d5fdbef38e04131b4656 labels: app.kubernetes.io/name: nvsentinel app.kubernetes.io/instance: nvsentinel spec: serviceAccountName: platform-connectors containers: - name: platform-connector image: "ghcr.io/nvidia/nvsentinel/platform-connectors:v1.9.0" imagePullPolicy: IfNotPresent securityContext: runAsUser: 0 capabilities: drop: - ALL ports: - name: metrics containerPort: 2112 startupProbe: httpGet: path: /healthz port: metrics periodSeconds: 10 timeoutSeconds: 5 failureThreshold: 30 livenessProbe: httpGet: path: /healthz port: metrics periodSeconds: 20 timeoutSeconds: 5 failureThreshold: 3 readinessProbe: httpGet: path: /healthz port: metrics periodSeconds: 10 timeoutSeconds: 3 failureThreshold: 3 args: - "--config=/etc/config/config.json" - "--metrics-port=2112" - "--mongo-client-cert-mount-path=/etc/ssl/mongo-client" - "--socket=/var/run/nvsentinel.sock" resources: limits: cpu: 200m memory: 512Mi requests: cpu: 200m memory: 512Mi volumeMounts: - name: var-run-vol mountPath: /var/run - name: platform-connector-configmap mountPath: /etc/config/ - name: mongo-app-client-cert mountPath: /etc/ssl/mongo-client readOnly: true env: - name: NODE_NAME valueFrom: fieldRef: apiVersion: v1 fieldPath: spec.nodeName - name: POD_NAME valueFrom: fieldRef: fieldPath: metadata.name - name: LOG_LEVEL value: "info" # App name for connection identification in logs and currentOp - name: APP_NAME value: "platform-connector" - name: MONGODB_CLIENT_CERT_MOUNT_PATH value: /etc/ssl/mongo-client envFrom: - configMapRef: name: mongodb-config optional: true volumes: - name: var-run-vol hostPath: path: /var/run/nvsentinel type: DirectoryOrCreate - name: platform-connector-configmap configMap: name: platform-connectors - name: mongo-app-client-cert secret: secretName: mongo-app-client-cert-secret items: - key: tls.crt path: tls.crt - key: tls.key path: tls.key - key: ca.crt path: ca.crt optional: true tolerations: - operator: Exists --- # Source: nvsentinel/charts/labeler/templates/deployment.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. apiVersion: apps/v1 kind: Deployment metadata: name: labeler labels: helm.sh/chart: labeler-0.1.0 app.kubernetes.io/name: labeler app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "1.16.0" app.kubernetes.io/managed-by: Helm annotations: argocd.argoproj.io/sync-wave: "0" spec: replicas: 1 selector: matchLabels: app.kubernetes.io/name: labeler app.kubernetes.io/instance: nvsentinel template: metadata: labels: helm.sh/chart: labeler-0.1.0 app.kubernetes.io/name: labeler app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "1.16.0" app.kubernetes.io/managed-by: Helm spec: serviceAccountName: labeler containers: - name: labeler image: "ghcr.io/nvidia/nvsentinel/labeler:v1.9.0" imagePullPolicy: IfNotPresent args: - "--metrics-port" - "2112" resources: limits: cpu: 500m memory: 256Mi requests: cpu: 100m memory: 128Mi ports: - name: metrics containerPort: 2112 livenessProbe: httpGet: path: /healthz port: metrics initialDelaySeconds: 15 periodSeconds: 20 timeoutSeconds: 5 failureThreshold: 3 readinessProbe: httpGet: path: /healthz port: metrics initialDelaySeconds: 5 periodSeconds: 10 timeoutSeconds: 3 failureThreshold: 3 env: - name: POD_NAME valueFrom: fieldRef: fieldPath: metadata.name - name: LOG_LEVEL value: "info" tolerations: - operator: Exists --- # Source: nvsentinel/templates/certmanager-postgresql.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. --- # Source: nvsentinel/templates/daemonset.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. --- # Source: nvsentinel/templates/networkpolicy.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. --- # Source: nvsentinel/templates/node-condition-cleanup-hook.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. --- # Source: nvsentinel/templates/podMonitor.yaml # Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. apiVersion: monitoring.coreos.com/v1 kind: PodMonitor metadata: name: nvsentinel labels: helm.sh/chart: nvsentinel-v1.9.0 app.kubernetes.io/name: nvsentinel app.kubernetes.io/instance: nvsentinel app.kubernetes.io/version: "v1.9.0" app.kubernetes.io/managed-by: Helm spec: selector: matchExpressions: - { key: "app.kubernetes.io/instance", operator: In, values: ["nvsentinel"] } namespaceSelector: matchNames: - nvsentinel podMetricsEndpoints: - port: metrics path: /metrics interval: 30s - port: cleanup-metrics path: /metrics interval: 30s