# Configurable Helm settings Type of document: Reference Product: F5 NGINX Instance Manager > Helm chart parameters, defaults, and examples for deploying F5 NGINX Instance Manager on Kubernetes. --- This reference lists Helm settings for F5 NGINX Instance Manager. The defaults below match the released `nim` chart **2.3.1**, for NGINX Instance Manager **2.23.1**. ## NGINX Instance Manager Helm chart settings {#helm-settings} For a direct installation of `nginx-stable/nim`, use the parameter names below without a `nim.` prefix. For example: ```yaml nmsClickhouse: mode: internal core: persistence: enabled: true ``` The Helm release name doesn't add a prefix to values. If you use this chart as a dependency, follow the parent chart's dependency name or alias. To check the defaults for your chart version, first [add the Helm repository](/nim/deploy/kubernetes/deploy-using-helm.md#add-repository). Then run: ```shell helm show values nginx-stable/nim --version ``` Replace `` with your chart version, such as `2.3.1`. See the [chart version table](/nim/deploy/kubernetes/deploy-using-helm.md#install-the-chart) for application versions. Use `helm upgrade` with `-f ` to apply changes. Replace `` with your values file path. Follow the [upgrade instructions](/nim/deploy/kubernetes/deploy-using-helm.md#upgrade-nginx-instance-manager) for the complete command. In the tables, `null` means unset, `""` means an empty string, `[]` means an empty list, and `{}` means an empty map. Resource requests are chart defaults, not sizing recommendations for every workload. ### General settings | Parameter | Description | Default | |---|---|---| | `adminPasswordHash` | Required admin password hash. Generate one with `openssl passwd -6 ''`, replacing `` with your password. | `null` | | `imagePullSecrets` | Kubernetes Secrets for pulling images from a private registry. | `[]` | | `nameOverride` | Name used by the chart's fullname helper when `fullnameOverride` is empty. | `nms` | | `fullnameOverride` | Full name for shared resources, including the ServiceAccount and Secrets. | `nms` | | `commonLabels` | Extra labels for resources that use the common-label helper. | `{}` | | `podLabels` | Extra pod labels for ClickHouse, API Gateway, Core, DPM, Ingestion, Integrations, Security Monitoring, and Vault. | `{}` | | `podAnnotations` | Extra pod annotations. This optional setting has no entry in the default values file. | Unset | | `serviceAccount.annotations` | Extra ServiceAccount annotations. | `{}` | | `serviceAccount.labels` | Extra ServiceAccount labels. | `{}` | | `serviceAccount.name` | ServiceAccount name used by pods. If you change `fullnameOverride`, set this to the same name or an existing ServiceAccount. | `nms` (template default) | | `openshift.enabled` | Use the chart's OpenShift settings. | `false` | | `global.utility` | Create the utility Deployment with zero replicas for maintenance tasks. | `false` | | `global.nmsModules` | Advanced module configuration supplied by module charts. | `{}` | | `nimLivenessAddress` | Address for internal liveness checks. | `apigw` | | `log.encoding` | Log format: `console` or `json`. | `console` | | `log.level` | Log level: `debug`, `info`, `warn`, `error`, `panic`, or `fatal`. | `error` | | `agent.secure` | Require agent client-certificate checks for the gateway's agent gRPC endpoints. | `false` | | `networkPolicies.enabled` | Create the chart's NetworkPolicies. | `true` | | `networkPolicies.apigw.ingressFrom` | NetworkPolicy peers allowed to reach the API Gateway. See the example below. | `[{ipBlock: {cidr: "0.0.0.0/0"}}]` | Use Kubernetes `namespaceSelector` and `podSelector` peers when you need to allow an ingress controller in another namespace. For example: ```yaml networkPolicies: apigw: ingressFrom: - namespaceSelector: matchLabels: kubernetes.io/metadata.name: ingress-nginx podSelector: matchLabels: app.kubernetes.io/name: ingress-nginx ``` Replace the namespace and pod labels with those of your ingress controller. Use Helm's `--namespace` option to select the deployment namespace. ### ClickHouse | Parameter | Description | Default | |---|---|---| | `nmsClickhouse.mode` | `internal` deploys ClickHouse, `external` uses an existing server, and `disabled` selects lightweight mode without metrics. | `internal` | | `nmsClickhouse.fullnameOverride` | Name for ClickHouse resources, including its Service. | `clickhouse` | | `nmsClickhouse.image.repository` | ClickHouse image repository. | `mirror.gcr.io/clickhouse/clickhouse-server` | | `nmsClickhouse.image.tag` | ClickHouse image tag. | `24.8.12.28-alpine` | | `nmsClickhouse.image.pullPolicy` | Image pull policy. | `IfNotPresent` | | `nmsClickhouse.auth.username` | ClickHouse username. | `default` | | `nmsClickhouse.auth.password` | ClickHouse password when no existing Secret is set. | `""` | | `nmsClickhouse.auth.existingSecret` | Existing Secret that contains the ClickHouse password. | `""` | | `nmsClickhouse.auth.existingSecretKey` | Password key in the existing Secret. | `""` | | `nmsClickhouse.service.rpcPort` | Native protocol port. | `9000` | | `nmsClickhouse.service.httpPort` | HTTP port exposed by the container and Service. | `8123` | | `nmsClickhouse.resources.requests.cpu` | CPU request. | `120m` | | `nmsClickhouse.resources.requests.memory` | Memory request. | `1Gi` | | `nmsClickhouse.persistence.enabled` | Use persistent storage for ClickHouse data. | `true` | | `nmsClickhouse.persistence.existingClaim` | Existing persistent volume claim (PVC). In chart 2.3.1, also set `nmsClickhouse.fullnameOverride` to this claim name. | `null` | | `nmsClickhouse.persistence.storageClass` | Storage class. An empty value uses the cluster default. Use `-` to request no storage class. | `""` | | `nmsClickhouse.persistence.volumeName` | Persistent volume to bind to the generated claim. | `""` | | `nmsClickhouse.persistence.accessMode` | Claim access mode. | `ReadWriteOnce` | | `nmsClickhouse.persistence.size` | Claim size. | `1Gi` | | `nmsClickhouse.tolerations` | Pod tolerations. See [Taints and tolerations](#kubernetes-taints-and-tolerations). | `[]` | | `externalClickhouse.address` | External ClickHouse host and native protocol port. Required when `nmsClickhouse.mode` is `external`. | `null` | | `externalClickhouse.auth.username` | External ClickHouse username. | `null` | | `externalClickhouse.auth.password` | External ClickHouse password when no existing Secret is set. | `""` | | `externalClickhouse.auth.existingSecret` | Existing Secret that contains the external ClickHouse password. | `""` | | `externalClickhouse.auth.existingSecretKey` | Password key in the existing Secret. | `""` | Set both `auth.existingSecret` and `auth.existingSecretKey` to use a password Secret. The existing Secret takes precedence over `auth.password`. The older `user` and `password` fields outside `auth` don't apply to chart 2.3.1. ### API Gateway | Parameter | Description | Default | |---|---|---| | `apigw.name` | Name for API Gateway resources, including its Service. | `apigw` | | `apigw.image.repository` | API Gateway image repository. | `apigw` | | `apigw.image.tag` | API Gateway image tag. | `2.23.1` | | `apigw.image.pullPolicy` | Image pull policy. | `IfNotPresent` | | `apigw.tlsSecret` | Existing `kubernetes.io/tls` Secret with `tls.crt` and `tls.key`. An empty value uses a chart-generated certificate. | `""` | | `apigw.serverName` | NGINX `server_name`. Accepts one name or a space-separated list. An empty value uses `_`. | `""` | | `apigw.tlsReload.enabled` | Watch the mounted certificate and send `SIGHUP` to NGINX when it rotates. | `false` | | `apigw.service.type` | Service type: `ClusterIP`, `LoadBalancer`, or `NodePort`. | `ClusterIP` | | `apigw.service.externalIPs` | External IP addresses for the Service. | `[]` | | `apigw.service.httpsPort` | Service HTTPS port. | `443` | | `apigw.service.nodePortHttps` | Fixed NodePort when the Service type is `NodePort`. If unset, Kubernetes allocates the port. | `null` | | `apigw.resources.requests.cpu` | CPU request. | `100m` | | `apigw.resources.requests.memory` | Memory request. | `256Mi` | | `apigw.tolerations` | Pod tolerations. | `[]` | | `apigw.nginxWorkerConnections` | NGINX `worker_connections`. | `1024` (template default) | | `apigw.nginxWorkerRlimitNofile` | NGINX `worker_rlimit_nofile`. | `4096` (template default) | | `replicaCount` | API Gateway replica count only. If unset, Kubernetes defaults to one replica. | Unset | The container listens on port `8443`. The Service forwards its HTTPS port to this container port. `apigw.container.port.https` and `apigw.service.name` don't control these settings in chart 2.3.1. For certificate setup, see [Use external TLS certificates](/nim/deploy/kubernetes/configure-external-certs.md). For worker tuning, see [Optimize NGINX proxy gateway](/nim/system-configuration/configure-gateway.md#update-worker-settings-with-helm). ### Core | Parameter | Description | Default | |---|---|---| | `core.name` | Core resource name. | `core` | | `core.image.repository` | Image repository. | `core` | | `core.image.tag` | Image tag. | `2.23.1` | | `core.image.pullPolicy` | Image pull policy. | `IfNotPresent` | | `core.container.port.http` | Container HTTP port. | `8033` | | `core.container.port.db` | Container database port. | `7891` | | `core.container.port.grpc` | Container gRPC port. | `8038` | | `core.service.httpPort` | Service HTTP port. | `8033` | | `core.service.grpcPort` | Service gRPC port. | `8038` | | `core.resources.requests.cpu` | CPU request. | `10m` | | `core.resources.requests.memory` | Memory request. | `512Mi` | | `core.persistence.enabled` | Use persistent storage. | `true` | | `core.persistence.claims` | Claims for dqlite and secrets. | See [Core storage](#nim-dqlite-storage-configuration) | | `core.persistence.storageClass` | Storage class. Empty uses the cluster default. Use `-` for no storage class. | `""` | | `core.persistence.volumeName` | Persistent volume to bind to generated claims. | `""` | | `core.tolerations` | Pod tolerations. | `[]` | ### Dataplane Manager | Parameter | Description | Default | |---|---|---| | `dpm.name` | Dataplane Manager resource name. | `dpm` | | `dpm.image.repository` | Image repository. | `dpm` | | `dpm.image.tag` | Image tag. | `2.23.1` | | `dpm.image.pullPolicy` | Image pull policy. | `IfNotPresent` | | `dpm.container.port.http` | Container HTTP port. | `8034` | | `dpm.container.port.grpc` | Container gRPC port. | `8036` | | `dpm.container.port.nats` | Container NATS port. | `9100` | | `dpm.container.port.db` | Container database port. | `7890` | | `dpm.service.httpPort` | Service HTTP port. | `8034` | | `dpm.service.grpcPort` | Service gRPC port. | `8036` | | `dpm.service.natsPort` | Service NATS port. | `9100` | | `dpm.resources.requests.cpu` | CPU request. | `100m` | | `dpm.resources.requests.memory` | Memory request. | `512Mi` | | `dpm.persistence.enabled` | Use persistent storage. | `true` | | `dpm.persistence.claims` | Claims for dqlite and NATS data. | See [DPM storage](#nim-nats-storage-configuration) | | `dpm.persistence.storageClass` | Storage class. Empty uses the cluster default. Use `-` for no storage class. | `""` | | `dpm.persistence.volumeName` | Persistent volume to bind to generated claims. | `""` | | `dpm.tolerations` | Pod tolerations. | `[]` | ### Ingestion | Parameter | Description | Default | |---|---|---| | `ingestion.name` | Ingestion resource name. | `ingestion` | | `ingestion.image.repository` | Image repository. | `ingestion` | | `ingestion.image.tag` | Image tag. | `2.23.1` | | `ingestion.image.pullPolicy` | Image pull policy. | `IfNotPresent` | | `ingestion.replicaCount` | Replica count. | `1` | | `ingestion.container.port.http` | Container HTTP port. | `8040` | | `ingestion.container.port.grpc` | Container gRPC port. | `8035` | | `ingestion.service.httpPort` | Service HTTP port. | `8040` | | `ingestion.service.grpcPort` | Service gRPC port. | `8035` | | `ingestion.resources.requests.cpu` | CPU request. | `100m` | | `ingestion.resources.requests.memory` | Memory request. | `512Mi` | | `ingestion.tolerations` | Pod tolerations. | `[]` | ### Integrations | Parameter | Description | Default | |---|---|---| | `integrations.name` | Integrations resource name. | `integrations` | | `integrations.image.repository` | Image repository. | `integrations` | | `integrations.image.tag` | Image tag. | `2.23.1` | | `integrations.image.pullPolicy` | Image pull policy. | `IfNotPresent` | | `integrations.container.modeOfOperation` | Licensing mode: `connected` or `disconnected`. | `connected` | | `integrations.container.port.http` | Container HTTP port. | `8037` | | `integrations.container.port.db` | Container database port. | `7892` | | `integrations.container.port.licenseDB` | License database port in chart 2.3.1. | `7893` | | `integrations.service.httpPort` | Service HTTP port. | `8037` | | `integrations.resources.requests.cpu` | CPU request. | `100m` | | `integrations.resources.requests.memory` | Memory request. | `512Mi` | | `integrations.persistence.enabled` | Use persistent storage. | `true` | | `integrations.persistence.claims` | Claims for dqlite and proxy certificates. | See [Integrations storage](#integrations-storage-configuration) | | `integrations.persistence.storageClass` | Storage class. Empty uses the cluster default. Use `-` for no storage class. | `""` | | `integrations.persistence.volumeName` | Persistent volume to bind to generated claims. | `""` | | `integrations.tolerations` | Pod tolerations. | `[]` | ### Security Monitoring | Parameter | Description | Default | |---|---|---| | `secmon.name` | Security Monitoring resource name. | `secmon` | | `secmon.image.repository` | Image repository. | `secmon` | | `secmon.image.tag` | Image tag. | `2.23.1` | | `secmon.image.pullPolicy` | Image pull policy. | `IfNotPresent` | | `secmon.replicaCount` | Replica count. | `1` | | `secmon.container.port.http` | Container HTTP port. | `8039` | | `secmon.service.httpPort` | Service HTTP port. | `8039` | | `secmon.resources.requests.cpu` | CPU request. | `100m` | | `secmon.resources.requests.memory` | Memory request. | `512Mi` | | `secmon.tolerations` | Pod tolerations. | `[]` | ### Utility Set `global.utility` to `true` to create the utility Deployment. It starts with zero replicas. See [Back up and restore](/nim/admin-guide/backup-and-recovery.md) for maintenance procedures. | Parameter | Description | Default | |---|---|---| | `utility.name` | Utility resource name. | `utility` | | `utility.image.repository` | Image repository. | `utility` | | `utility.image.tag` | Image tag. | `2.23.1` | | `utility.image.pullPolicy` | Image pull policy. | `IfNotPresent` | | `utility.tolerations` | Pod tolerations. | `[]` | ### Labels and security contexts For each of `nmsClickhouse`, `apigw`, `core`, `dpm`, `ingestion`, `integrations`, and `secmon`, you can set these labels. Replace `` with a service name. | Parameter | Description | Default | |---|---|---| | `.service.labels` | Extra Service labels. | `{}` | | `.pod.labels` | Extra pod labels. | `{}` | The security context fields below set the pod user, group, and volume group. Replace `` with a service from the next table. | Parameter | Description | |---|---| | `.securityContext.runAsUser` | User ID for the pod's processes. | | `.securityContext.runAsGroup` | Primary group ID. | | `.securityContext.fsGroup` | Group ID for mounted volumes. | | Service | `runAsUser` | `runAsGroup` | `fsGroup` | |---|---|---|---| | `nmsClickhouse` | `101` | `101` | `101` | | `apigw` | `101` | `101` | `101` | | `core` | `1000` | `1000` | `1000` | | `dpm` | `1000` | `1000` | `1000` | | `ingestion` | `1000` | `1000` | `1000` | | `integrations` | `1000` | `1000` | `1000` | | `secmon` | `1000` | `1000` | `1000` | | `utility` | `1000` | `1000` | `1000` | When `openshift.enabled` is `true`, the chart uses the following settings instead of the per-service values. Leave them unset for OpenShift to assign IDs. | Parameter | Default | |---|---| | `openshift.securityContext.runAsUser` | `null` | | `openshift.securityContext.runAsGroup` | `null` | | `openshift.securityContext.fsGroup` | `null` | ### External service certificates Use these settings to supply certificates for internal mutual TLS (mTLS). Follow [Use external TLS certificates](/nim/deploy/kubernetes/configure-external-certs.md) for Secret contents and setup. | Parameter | Description | Default | |---|---|---| | `externalCerts.ca.enabled` | Use a pre-created CA Secret instead of generating one. | `false` | | `externalCerts.ca.secretName` | CA Secret name override. | `""` | | `externalCerts.core.enabled` | Use pre-created Core certificates. | `false` | | `externalCerts.core.secretName` | Core Secret name override. | `""` | | `externalCerts.dpm.enabled` | Use pre-created DPM certificates. | `false` | | `externalCerts.dpm.secretName` | DPM Secret name override. | `""` | | `externalCerts.ingestion.enabled` | Use pre-created Ingestion certificates. | `false` | | `externalCerts.ingestion.secretName` | Ingestion Secret name override. | `""` | | `externalCerts.integrations.enabled` | Use pre-created Integrations certificates. | `false` | | `externalCerts.integrations.secretName` | Integrations Secret name override. | `""` | | `externalCerts.secmon.enabled` | Use pre-created Security Monitoring certificates. | `false` | | `externalCerts.secmon.secretName` | Security Monitoring Secret name override. | `""` | | `externalCerts.apigw.enabled` | Use pre-created API Gateway client certificates. | `false` | | `externalCerts.apigw.secretName` | API Gateway client Secret name override. | `""` | If a Secret name is empty, the chart derives it from its fullname. With the default fullname `nms`, the names are `nms-ca`, `nms-core-certs`, `nms-dpm-certs`, `nms-ingestion-certs`, `nms-integrations-certs`, `nms-secmon-certs`, and `nms-apigw-client-certs`. ### Forward proxy | Parameter | Description | Default | |---|---|---| | `proxyConfig.proxyEnable` | Use a forward proxy. | `false` | | `proxyConfig.proxyHost` | Proxy hostname. | `""` | | `proxyConfig.proxyPort` | Proxy port. | `3128` | | `proxyConfig.proxyProtocol` | Proxy protocol: `http` or `https`. | `http` | | `proxyConfig.proxyAuthRequired` | Use basic authentication with the proxy. | `false` | | `proxyConfig.proxyUsername` | Basic authentication username. | `""` | | `proxyConfig.proxyPassword` | Basic authentication password. | `""` | | `proxyConfig.proxySslVerify` | Verify the proxy certificate. In chart 2.3.1, an explicit `false` also renders as `true`. | `true` | See [Use a third-party forward proxy](/nim/system-configuration/configure-forward-proxy.md) for proxy requirements. ### Vault Use `externalVault.address` for an existing Vault server. See [Use Vault for storing secrets](/nim/system-configuration/configure-vault.md) for the required token and server setup. The bundled `nmsVault` settings are for development only. In chart 2.3.1, setting `nmsVault.enabled` to `true` fails template rendering. The Vault label templates use a different values path. Use an existing Vault server for this version. | Parameter | Description | Default | |---|---|---| | `externalVault.address` | External Vault address. | `null` | | `nmsVault.enabled` | Request the development Vault Deployment. | `false` | | `nmsVault.fullnameOverride` | Name for development Vault resources. | `vault` | | `nmsVault.image.repository` | Development Vault image repository. | `vault` | | `nmsVault.image.tag` | Development Vault image tag. | `1.12.2` | | `nmsVault.image.pullPolicy` | Image pull policy. | `IfNotPresent` | | `nmsVault.service.rpcPort` | Vault port. | `8200` | ## Core dqlite storage configuration {#nim-dqlite-storage-configuration} Core uses a `500Mi` dqlite claim and a `128Mi` secrets claim. Use the complete array when you override `core.persistence.claims`: ```yaml core: persistence: claims: - name: dqlite existingClaim: null size: 500Mi accessMode: ReadWriteOnce - name: secrets existingClaim: null size: 128Mi accessMode: ReadWriteOnce ``` ## Core secrets storage configuration {#nim-secrets-storage-configuration} The `secrets` entry in the [Core claim array](#nim-dqlite-storage-configuration) stores secrets. To reuse a claim, set that entry's `existingClaim` to its name. Create the claim before installation. ## DPM dqlite and NATS storage configuration {#nim-nats-storage-configuration} DPM uses separate `1Gi` claims for dqlite and NATS data: ```yaml dpm: persistence: claims: - name: dqlite existingClaim: null size: 1Gi accessMode: ReadWriteOnce - name: nats-streaming existingClaim: null size: 1Gi accessMode: ReadWriteOnce ``` The `nats-streaming` claim name remains in the chart for the NATS data volume. ## Integrations storage configuration Integrations uses a `1Gi` dqlite claim and a `128Mi` proxy-certificate claim: ```yaml integrations: persistence: claims: - name: dqlite existingClaim: null size: 1Gi accessMode: ReadWriteOnce - name: proxy-certs existingClaim: null size: 128Mi accessMode: ReadWriteOnce ``` For DPM and Integrations, set `existingClaim` on each entry to reuse a pre-created claim. Helm replaces claim arrays rather than merging individual entries, so include both claims when you override an array. ## Kubernetes taints and tolerations Set tolerations under the service name. This example lets the API Gateway pod remain bound to an unreachable node for 60 seconds before eviction: ```yaml apigw: tolerations: - key: node.kubernetes.io/unreachable operator: Exists effect: NoExecute tolerationSeconds: 60 ``` For more information, see Kubernetes [Taints and Tolerations](https://kubernetes.io/docs/concepts/scheduling-eviction/taint-and-toleration/). ## Earlier chart versions Starting with NGINX Instance Manager 2.20.0, the chart name is `nginx-stable/nim`. Its first chart version is `2.0.0`. For NGINX Instance Manager 2.19.x, use `nginx-stable/nms-hybrid`. Direct installations use root-level values. For older `nginx-stable/nms` installations, values for the dependency use the `nms-hybrid` prefix. These older charts use `nmsClickhouse.enabled` instead of `nmsClickhouse.mode`. Check your chart's values and follow the [version-specific installation instructions](/nim/deploy/kubernetes/deploy-using-helm.md#helm-deployment-for-nginx-instance-manager-219) before applying settings from this reference.