You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
5.6 KiB
5.6 KiB
NeMo Guardrails Microservice Helm Chart
For deployment guide, see Admin Setup in the NeMo Microservices documentation.
Values
| Key | Type | Default | Description |
|---|---|---|---|
| affinity | object | {} |
Specifies affinity settings for the deployment. To learn more, refer to the Affinity and Anti-Affinity in the Kubernetes documentation. |
| autoscaling | object | {"enabled":false,"maxReplicas":100,"minReplicas":1,"targetCPUUtilizationPercentage":80} |
Specifies autoscaling configurations for the deployment. |
| autoscaling.enabled | bool | false |
Whether to enable horizontal pod autoscaler. |
| autoscaling.maxReplicas | int | 100 |
The maximum number of replicas for the deployment. |
| autoscaling.minReplicas | int | 1 |
The minimum number of replicas for the deployment. |
| autoscaling.targetCPUUtilizationPercentage | int | 80 |
The target CPU utilization percentage. |
| configStore.nfs.enabled | bool | false |
Whether to enable the use of an NFS persistent volume for the configuration store. |
| configStore.nfs.mountPath | string | "/config-store" |
The path where the NFS volume will be mounted inside the container. |
| configStore.nfs.path | string | "/path/to/nfs/share" |
The path to the root of the Configuration Store folder. |
| configStore.nfs.server | string | "nfs-server.example.com" |
The address of the NFS server. |
| configStore.nfs.storageClass | string | "standard" |
The storage class for the PV and PVC. |
| env | object | {"CONFIG_STORE_PATH":"/app/services/guardrails/config-store","DEFAULT_CONFIG_ID":"default","DEFAULT_LLM_PROVIDER":"nim","DEMO":"False","GUARDRAILS_HOST":"0.0.0.0","GUARDRAILS_PORT":"7331","NEMO_GUARDRAILS_SERVER_ALLOWED_ORIGINS":"*","NEMO_GUARDRAILS_SERVER_ENABLE_CORS":"False","NIM_ENDPOINT_URL":"https://integrate.api.nvidia.com/v1"} |
Environment variables for the container. |
| fsGroup | string | "" |
Specifies the file system owner group id. |
| fullnameOverride | string | "" |
Overrides the full chart name. |
| image.pullPolicy | string | "IfNotPresent" |
The image pull policy for the NeMo Guardrails container image. |
| image.repository | string | "nvcr.io/nvidia/nemo-microservices/guardrails" |
The repository location of the NeMo Guardrails container image. |
| image.tag | string | "" |
The tag of the NeMo Guardrails container image. |
| imagePullSecrets | list | [{"name":"nvcrimagepullsecret"}] |
Specifies the list of secret names that are needed for the main container and any init containers. |
| ingress.annotations | object | {} |
Additional annotations for the ingress resource. |
| ingress.className | string | "" |
The class name for the ingress resource. |
| ingress.enabled | bool | false |
Whether to enable the ingress resource. |
| ingress.hosts | list | [] | A list of maps, each containing the keys host and paths for the ingress resource. You must specify a list for configuring ingress for the microservice. |
| ingress.tls | list | [] |
TLS configuration for the ingress resource. |
| livenessProbe.httpGet | object | {"path":"/v1/health","port":7331} |
The HTTP GET request to use for the liveness probe. |
| livenessProbe.initialDelaySeconds | int | 5 |
The initial delay seconds for the liveness probe. |
| livenessProbe.timeoutSeconds | int | 30 |
The timeout in seconds for the liveness probe. |
| nameOverride | string | "" |
Overrides the chart name. |
| nodeSelector | object | {} |
Specifies labels to ensure that the NeMo Guardrails microservice is deployed only on certain nodes. To learn more, refer to the Node Selector in the Kubernetes documentation. |
| podAnnotations | object | {} |
Specifies additional annotations to the main deployment pods. |
| podLabels | object | {} |
Specifies additional labels to the main deployment pods. |
| podSecurityContext | object | {} |
Specifies privilege and access control settings for the pod. |
| readinessProbe.httpGet | object | {"path":"/v1/health","port":7331} |
The HTTP GET request to use for the readiness probe. |
| readinessProbe.initialDelaySeconds | int | 5 |
The initial delay seconds for the readiness probe. |
| readinessProbe.timeoutSeconds | int | 30 |
The timeout in seconds for the readiness probe. |
| replicaCount | int | 1 |
The number of replicas for the NeMo Guardrails microservice deployment. |
| resources | object | {} |
Specifies resource configurations for the deployment. |
| runAsGroup | string | "" |
Specifies the group ID for the pod. |
| securityContext | object | {} |
Specifies security context for the container. |
| service | object | {"port":7331,"type":"ClusterIP"} |
Specifies the service type and the port for the deployment. |
| serviceAccount.annotations | object | {} |
Annotations to be added to the service account. |
| serviceAccount.automount | bool | true |
Whether to automatically mount the service account token. |
| serviceAccount.create | bool | true |
Whether to create a service account for the NeMo Guardrails microservice. |
| serviceAccount.name | string | "" |
The name of the service account to use. |
| tolerations | list | [] |
Specifies tolerations for pod assignment. To learn more, refer to the Taint and Toleration in the Kubernetes documentation. |