# Copyright VMware, Inc. # SPDX-License-Identifier: APACHE-2.0 ## @section Global parameters ## Global Docker image parameters ## Please, note that this will override the image parameters, including dependencies, configured to use the global value ## Current available global Docker image parameters: imageRegistry, imagePullSecrets and storageClass ## ## @param global.imageRegistry Global Docker image registry ## @param global.imagePullSecrets Global Docker registry secret names as an array ## @param global.storageClass Global StorageClass for Persistent Volume(s) ## global: imageRegistry: "" ## E.g. ## imagePullSecrets: ## - myRegistryKeySecretName ## imagePullSecrets: [] storageClass: "" ## @section Common parameters ## ## @param kubeVersion Override Kubernetes version ## kubeVersion: "" ## @param nameOverride String to partially override common.names.fullname template (will maintain the release name) ## nameOverride: "" ## @param commonLabels Labels to add to all deployed objects ## commonLabels: {} ## @param commonAnnotations Annotations to add to all deployed objects ## commonAnnotations: {} ## @param fullnameOverride String to fully override common.names.fullname template ## fullnameOverride: "" ## @param extraDeploy Array of extra objects to deploy with the release ## extraDeploy: [] ## Enable diagnostic mode in the deployment ## diagnosticMode: ## @param diagnosticMode.enabled Enable diagnostic mode (all probes will be disabled and the command will be overridden) ## enabled: false ## @param diagnosticMode.command Command to override all containers in the deployment ## command: - sleep ## @param diagnosticMode.args Args to override all containers in the deployment ## args: - infinity ## @section PyTorch parameters ## ## Bitnami PyTorch image version ## ref: https://hub.docker.com/r/bitnami/pytorch/tags/ ## @param image.registry [default: REGISTRY_NAME] PyTorch image registry ## @param image.repository [default: REPOSITORY_NAME/pytorch] PyTorch image repository ## @skip image.tag PyTorch image tag (immutable tags are recommended) ## @param image.digest PyTorch image digest in the way sha256:aa.... Please note this parameter, if set, will override the tag ## @param image.pullPolicy PyTorch image pull policy ## @param image.pullSecrets Specify docker-registry secret names as an array ## image: registry: docker.io repository: bitnami/pytorch tag: 2.1.2-debian-11-r1 digest: "" ## Specify a imagePullPolicy ## Defaults to 'Always' if image tag is 'latest', else set to 'IfNotPresent' ## ref: https://kubernetes.io/docs/concepts/containers/images/#pre-pulled-images ## pullPolicy: IfNotPresent ## Optionally specify an array of imagePullSecrets. ## Secrets must be manually created in the namespace. ## ref: https://kubernetes.io/docs/tasks/configure-pod-container/pull-image-private-registry/ ## Example: ## pullSecrets: ## - myRegistryKeySecretName ## pullSecrets: [] ## @param worldSize Number of nodes that will run the code ## WORLD_SIZE will be set to this value ## worldSize: 1 ## port PyTorch master port. DEPRECATED Use containerPorts.pytorch instead ## ## port: 49875 ## @param containerPorts.pytorch PyTorch master port. `MASTER_PORT` will be set to this value ## containerPorts: pytorch: 49875 ## Configure extra options for liveness probe ## ref: https://kubernetes.io/docs/tasks/configure-pod-container/configure-liveness-readiness-probes/#configure-probes ## @param livenessProbe.enabled Enable livenessProbe ## @param livenessProbe.initialDelaySeconds Initial delay seconds for livenessProbe ## @param livenessProbe.periodSeconds Period seconds for livenessProbe ## @param livenessProbe.timeoutSeconds Timeout seconds for livenessProbe ## @param livenessProbe.failureThreshold Failure threshold for livenessProbe ## @param livenessProbe.successThreshold Success threshold for livenessProbe ## livenessProbe: enabled: true initialDelaySeconds: 5 periodSeconds: 5 timeoutSeconds: 5 successThreshold: 1 failureThreshold: 5 ## Configure extra options for readiness probe ## ref: https://kubernetes.io/docs/tasks/configure-pod-container/configure-liveness-readiness-probes/#configure-probes ## @param readinessProbe.enabled Enable readinessProbe ## @param readinessProbe.initialDelaySeconds Initial delay seconds for readinessProbe ## @param readinessProbe.periodSeconds Period seconds for readinessProbe ## @param readinessProbe.timeoutSeconds Timeout seconds for readinessProbe ## @param readinessProbe.failureThreshold Failure threshold for readinessProbe ## @param readinessProbe.successThreshold Success threshold for readinessProbe ## readinessProbe: enabled: true initialDelaySeconds: 5 periodSeconds: 5 timeoutSeconds: 3 successThreshold: 1 failureThreshold: 5 ## Configure extra options for startup probe ## ref: https://kubernetes.io/docs/tasks/configure-pod-container/configure-liveness-startup-probes/#configure-probes ## @param startupProbe.enabled Enable startupProbe ## @param startupProbe.initialDelaySeconds Initial delay seconds for startupProbe ## @param startupProbe.periodSeconds Period seconds for startupProbe ## @param startupProbe.timeoutSeconds Timeout seconds for startupProbe ## @param startupProbe.failureThreshold Failure threshold for startupProbe ## @param startupProbe.successThreshold Success threshold for startupProbe ## startupProbe: enabled: true initialDelaySeconds: 5 periodSeconds: 5 timeoutSeconds: 3 successThreshold: 1 failureThreshold: 5 ## @param customLivenessProbe Custom livenessProbe that overrides the default one ## customLivenessProbe: {} ## @param customReadinessProbe Custom readinessProbe that overrides the default one ## customReadinessProbe: {} ## @param customStartupProbe Custom startupProbe that overrides the default one ## customStartupProbe: {} ## Pod Security Context ## ref: https://kubernetes.io/docs/tasks/configure-pod-container/security-context/ ## securityContext.enabled DEPRECATED Use podSecurityContext.enabled instead ## securityContext.fsGroup DEPRECATED Use podSecurityContext.fsGroup instead ## securityContext.runAsUser DEPRECATED Use podSecurityContext.runAsUser instead ## securityContext: ## enabled: true ## fsGroup: 1001 ## runAsUser: 1001 ## @param podSecurityContext.enabled Enabled Pytorch pods' Security Context ## @param podSecurityContext.fsGroupChangePolicy Set filesystem group change policy ## @param podSecurityContext.sysctls Set kernel settings using the sysctl interface ## @param podSecurityContext.supplementalGroups Set filesystem extra groups ## @param podSecurityContext.fsGroup Set Pytorch pods' Security Context fsGroup ## podSecurityContext: enabled: true fsGroupChangePolicy: Always sysctls: [] supplementalGroups: [] fsGroup: 1001 ## Configure Container Security Context ## ref: https://kubernetes.io/docs/tasks/configure-pod-container/security-context/#set-the-security-context-for-a-container ## @param containerSecurityContext.enabled Enabled containers' Security Context ## @param containerSecurityContext.seLinuxOptions [object,nullable] Set SELinux options in container ## @param containerSecurityContext.runAsUser Set containers' Security Context runAsUser ## @param containerSecurityContext.runAsNonRoot Set container's Security Context runAsNonRoot ## @param containerSecurityContext.privileged Set container's Security Context privileged ## @param containerSecurityContext.readOnlyRootFilesystem Set container's Security Context readOnlyRootFilesystem ## @param containerSecurityContext.allowPrivilegeEscalation Set container's Security Context allowPrivilegeEscalation ## @param containerSecurityContext.capabilities.drop List of capabilities to be dropped ## @param containerSecurityContext.seccompProfile.type Set container's Security Context seccomp profile ## containerSecurityContext: enabled: true seLinuxOptions: null runAsUser: 1001 runAsNonRoot: true privileged: false readOnlyRootFilesystem: false allowPrivilegeEscalation: false capabilities: drop: ["ALL"] seccompProfile: type: "RuntimeDefault" ## Pytorch resource requests and limits ## ref: http://kubernetes.io/docs/concepts/configuration/manage-compute-resources-container/ ## @param resources.limits The resources limits for the Pytorch containers ## @param resources.requests The requested resources for the Pytorch containers ## resources: limits: {} requests: {} ## PyTorch configuration. This will be executed as: python [file] [args] ## @param entrypoint.file Main entrypoint to your application ## @param entrypoint.args Args required by your entrypoint ## entrypoint: file: "" args: [] ## mode DEPRECATED. Use architecture instead ## ## mode: standalone ## @param architecture Run PyTorch in standalone or distributed mode. Possible values: `standalone`, `distributed` ## architecture: standalone ## @param automountServiceAccountToken Mount Service Account token in pod ## automountServiceAccountToken: false ## @param hostAliases Deployment pod host aliases ## https://kubernetes.io/docs/concepts/services-networking/add-entries-to-pod-etc-hosts-with-host-aliases/ ## hostAliases: [] ## @param command Override default container command (useful when using custom images) ## command: [] ## @param args Override default container args (useful when using custom images) ## args: [] ## @param podLabels Extra labels for Pytorch pods ## ref: https://kubernetes.io/docs/concepts/overview/working-with-objects/labels/ ## podLabels: {} ## @param podAnnotations Annotations for Pytorch pods ## ref: https://kubernetes.io/docs/concepts/overview/working-with-objects/annotations/ ## podAnnotations: {} ## @param existingConfigmap Config map that contains the files you want to load in PyTorch ## existingConfigmap: "" ## Enable in order to download files from git repository. ## cloneFilesFromGit: ## @param cloneFilesFromGit.enabled Enable in order to download files from git repository ## enabled: false ## @param cloneFilesFromGit.repository Repository that holds the files ## repository: "" ## @param cloneFilesFromGit.revision Revision from the repository to checkout ## revision: "" ## @param cloneFilesFromGit.extraVolumeMounts Add extra volume mounts for the Git container ## Useful to mount keys to connect through ssh. (normally used with extraVolumes) ## E.g: ## extraVolumeMounts: ## - name: ssh-dir ## mountPath: /.ssh/ ## extraVolumeMounts: [] ## @param podAffinityPreset Pod affinity preset. Ignored if `affinity` is set. Allowed values: `soft` or `hard` ## ref: https://kubernetes.io/docs/concepts/scheduling-eviction/assign-pod-node/#inter-pod-affinity-and-anti-affinity ## podAffinityPreset: "" ## @param podAntiAffinityPreset Pod anti-affinity preset. Ignored if `affinity` is set. Allowed values: `soft` or `hard` ## Ref: https://kubernetes.io/docs/concepts/scheduling-eviction/assign-pod-node/#inter-pod-affinity-and-anti-affinity ## podAntiAffinityPreset: soft ## Node affinity preset ## Ref: https://kubernetes.io/docs/concepts/scheduling-eviction/assign-pod-node/#node-affinity ## nodeAffinityPreset: ## @param nodeAffinityPreset.type Node affinity preset type. Ignored if `affinity` is set. Allowed values: `soft` or `hard` ## type: "" ## @param nodeAffinityPreset.key Node label key to match Ignored if `affinity` is set. ## E.g. ## key: "kubernetes.io/e2e-az-name" ## key: "" ## @param nodeAffinityPreset.values Node label values to match. Ignored if `affinity` is set. ## E.g. ## values: ## - e2e-az1 ## - e2e-az2 ## values: [] ## @param affinity Affinity for pod assignment. Evaluated as a template. ## Ref: https://kubernetes.io/docs/concepts/configuration/assign-pod-node/#affinity-and-anti-affinity ## Note: podAffinityPreset, podAntiAffinityPreset, and nodeAffinityPreset will be ignored when it's set ## affinity: {} ## @param nodeSelector Node labels for pod assignment. Evaluated as a template. ## ref: https://kubernetes.io/docs/concepts/scheduling-eviction/assign-pod-node/ ## nodeSelector: {} ## @param tolerations Tolerations for pod assignment. Evaluated as a template. ## ref: https://kubernetes.io/docs/concepts/configuration/taint-and-toleration/ ## tolerations: [] ## @param updateStrategy.type Pytorch statefulset strategy type ## ref: https://kubernetes.io/docs/concepts/workloads/controllers/statefulset/#update-strategies ## updateStrategy: ## StrategyType ## Can be set to RollingUpdate or OnDelete ## type: RollingUpdate ## @param podManagementPolicy Statefulset Pod management policy, it needs to be Parallel to be able to complete the cluster join ## Ref: https://kubernetes.io/docs/concepts/workloads/controllers/statefulset/#pod-management-policies ## podManagementPolicy: OrderedReady ## @param priorityClassName Pytorch pods' priorityClassName ## priorityClassName: "" ## @param topologySpreadConstraints Topology Spread Constraints for pod assignment spread across your cluster among failure-domains. Evaluated as a template ## Ref: https://kubernetes.io/docs/concepts/workloads/pods/pod-topology-spread-constraints/#spread-constraints-for-pods ## topologySpreadConstraints: [] ## @param schedulerName Name of the k8s scheduler (other than default) for Pytorch pods ## ref: https://kubernetes.io/docs/tasks/administer-cluster/configure-multiple-schedulers/ ## schedulerName: "" ## @param terminationGracePeriodSeconds Seconds Redmine pod needs to terminate gracefully ## ref: https://kubernetes.io/docs/concepts/workloads/pods/pod/#termination-of-pods ## terminationGracePeriodSeconds: "" ## @param lifecycleHooks for the Pytorch container(s) to automate configuration before or after startup ## lifecycleHooks: {} ## @param extraEnvVars Array with extra environment variables to add to Pytorch nodes ## e.g: ## extraEnvVars: ## - name: FOO ## value: "bar" ## extraEnvVars: [] ## @param extraEnvVarsCM Name of existing ConfigMap containing extra env vars for Pytorch nodes ## extraEnvVarsCM: "" ## @param extraEnvVarsSecret Name of existing Secret containing extra env vars for Pytorch nodes ## extraEnvVarsSecret: "" ## @param extraVolumes Optionally specify extra list of additional volumes for the Pytorch pod(s) ## extraVolumes: [] ## @param extraVolumeMounts Optionally specify extra list of additional volumeMounts for the Pytorch container(s) ## extraVolumeMounts: [] ## @param sidecars Add additional sidecar containers to the Pytorch pod(s) ## e.g: ## sidecars: ## - name: your-image-name ## image: your-image ## imagePullPolicy: Always ## ports: ## - name: portname ## containerPort: 1234 ## sidecars: [] ## @param initContainers Add additional init containers to the %%MAIN_CONTAINER_NAME%% pod(s) ## ref: https://kubernetes.io/docs/concepts/workloads/pods/init-containers/ ## e.g: ## initContainers: ## - name: your-image-name ## image: your-image ## imagePullPolicy: Always ## command: ['sh', '-c', 'echo "hello world"'] ## initContainers: [] ## Service Account ## ref: https://kubernetes.io/docs/tasks/configure-pod-container/configure-service-account/ ## serviceAccount: ## @param serviceAccount.create Enable creation of ServiceAccount for Pytorch pod ## create: true ## @param serviceAccount.name The name of the ServiceAccount to use. ## If not set and create is true, a name is generated using the common.names.fullname template ## name: "" ## @param serviceAccount.automountServiceAccountToken Allows auto mount of ServiceAccountToken on the serviceAccount created ## Can be set to false if pods using this serviceAccount do not need to use K8s API ## automountServiceAccountToken: false ## @param serviceAccount.annotations Additional custom annotations for the ServiceAccount ## annotations: {} ## @section Traffic Exposure Parameters ## ## Service for the scheduler node ## service: ## @param service.type Kubernetes service type ## type: ClusterIP ## service.port DEPRECATED Use service.ports.pytorch instead ## ## port: 49875 ## @param service.ports.pytorch Scheduler Service port ## ports: pytorch: 49875 ## Node ports to expose ## service.nodePort DEPRECATED Use service.nodePorts.pytorch instead ## ## nodePort: "" ## @param service.nodePorts.pytorch Node port for Pytorch ## NOTE: choose port between <30000-32767> ## nodePorts: pytorch: "" ## @param service.clusterIP Pytorch service Cluster IP ## e.g.: ## clusterIP: None ## clusterIP: "" ## @param service.loadBalancerIP Pytorch service Load Balancer IP ## ref: https://kubernetes.io/docs/concepts/services-networking/service/#type-loadbalancer ## loadBalancerIP: "" ## @param service.loadBalancerSourceRanges Pytorch service Load Balancer sources ## ref: https://kubernetes.io/docs/tasks/access-application-cluster/configure-cloud-provider-firewall/#restrict-access-for-loadbalancer-service ## e.g: ## loadBalancerSourceRanges: ## - 10.10.10.0/24 ## loadBalancerSourceRanges: [] ## @param service.externalTrafficPolicy Pytorch service external traffic policy ## ref http://kubernetes.io/docs/tasks/access-application-cluster/create-external-load-balancer/#preserving-the-client-source-ip ## externalTrafficPolicy: Cluster ## @param service.annotations Additional custom annotations for Pytorch service ## annotations: {} ## @param service.extraPorts Extra ports to expose in Pytorch service (normally used with the `sidecars` value) ## extraPorts: [] ## @param service.sessionAffinity Control where client requests go, to the same pod or round-robin ## Values: ClientIP or None ## ref: https://kubernetes.io/docs/concepts/services-networking/service/ ## sessionAffinity: None ## @param service.sessionAffinityConfig Additional settings for the sessionAffinity ## sessionAffinityConfig: ## clientIP: ## timeoutSeconds: 300 ## sessionAffinityConfig: {} ## Headless service properties ## headless: ## @param service.headless.annotations Annotations for the headless service. ## annotations: {} ## Network Policies ## Ref: https://kubernetes.io/docs/concepts/services-networking/network-policies/ ## networkPolicy: ## @param networkPolicy.enabled Specifies whether a NetworkPolicy should be created ## enabled: true ## @param networkPolicy.allowExternal Don't require server label for connections ## The Policy model to apply. When set to false, only pods with the correct ## server label will have network access to the ports server is listening ## on. When true, server will accept connections from any source ## (with the correct destination port). ## allowExternal: true ## @param networkPolicy.allowExternalEgress Allow the pod to access any range of port and all destinations. ## allowExternalEgress: true ## @param networkPolicy.extraIngress [array] Add extra ingress rules to the NetworkPolice ## e.g: ## extraIngress: ## - ports: ## - port: 1234 ## from: ## - podSelector: ## - matchLabels: ## - role: frontend ## - podSelector: ## - matchExpressions: ## - key: role ## operator: In ## values: ## - frontend extraIngress: [] ## @param networkPolicy.extraEgress [array] Add extra ingress rules to the NetworkPolicy ## e.g: ## extraEgress: ## - ports: ## - port: 1234 ## to: ## - podSelector: ## - matchLabels: ## - role: frontend ## - podSelector: ## - matchExpressions: ## - key: role ## operator: In ## values: ## - frontend ## extraEgress: [] ## @param networkPolicy.ingressNSMatchLabels [object] Labels to match to allow traffic from other namespaces ## @param networkPolicy.ingressNSPodMatchLabels [object] Pod labels to match to allow traffic from other namespaces ## ingressNSMatchLabels: {} ingressNSPodMatchLabels: {} ## @section Init Container Parameters ## ## Bitnami git image version ## ref: https://hub.docker.com/r/bitnami/git/tags/ ## @param git.registry [default: REGISTRY_NAME] Git image registry ## @param git.repository [default: REPOSITORY_NAME/git] Git image repository ## @skip git.tag Git image tag (immutable tags are recommended) ## @param git.digest Git image digest in the way sha256:aa.... Please note this parameter, if set, will override the tag ## @param git.pullPolicy Git image pull policy ## @param git.pullSecrets Specify docker-registry secret names as an array ## git: registry: docker.io repository: bitnami/git tag: 2.43.0-debian-11-r5 digest: "" pullPolicy: IfNotPresent ## Optionally specify an array of imagePullSecrets. ## Secrets must be manually created in the namespace. ## ref: https://kubernetes.io/docs/tasks/configure-pod-container/pull-image-private-registry/ ## Example: ## pullSecrets: ## - myRegistryKeySecretName ## pullSecrets: [] ## 'volumePermissions' init container parameters ## Changes the owner and group of the persistent volume mount point to runAsUser:fsGroup values ## based on the *podSecurityContext/*containerSecurityContext parameters ## volumePermissions: ## @param volumePermissions.enabled Enable init container that changes volume permissions in the data directory (for cases where the default k8s `runAsUser` and `fsUser` values do not work) ## enabled: false ## @param volumePermissions.image.registry [default: REGISTRY_NAME] Init container volume-permissions image registry ## @param volumePermissions.image.repository [default: REPOSITORY_NAME/os-shell] Init container volume-permissions image repository ## @skip volumePermissions.image.tag Init container volume-permissions image tag (immutable tags are recommended) ## @param volumePermissions.image.digest Init container volume-permissions image digest in the way sha256:aa.... Please note this parameter, if set, will override the tag ## @param volumePermissions.image.pullPolicy Init container volume-permissions image pull policy ## @param volumePermissions.image.pullSecrets Specify docker-registry secret names as an array ## image: registry: docker.io repository: bitnami/os-shell tag: 11-debian-11-r94 digest: "" pullPolicy: IfNotPresent ## Optionally specify an array of imagePullSecrets. ## Secrets must be manually created in the namespace. ## ref: https://kubernetes.io/docs/tasks/configure-pod-container/pull-image-private-registry/ ## Example: ## pullSecrets: ## - myRegistryKeySecretName ## pullSecrets: [] ## Init container' resource requests and limits ## ref: https://kubernetes.io/docs/concepts/configuration/manage-compute-resources-container/ ## We usually recommend not to specify default resources and to leave this as a conscious ## choice for the user. This also increases chances charts run on environments with little ## resources, such as Minikube. If you do want to specify resources, uncomment the following ## lines, adjust them as necessary, and remove the curly braces after 'resources:'. ## @param volumePermissions.resources.limits The resources limits for the container ## @param volumePermissions.resources.requests The requested resources for the container ## resources: ## Example: ## limits: ## cpu: 100m ## memory: 128Mi ## limits: {} ## Examples: ## requests: ## cpu: 100m ## memory: 128Mi ## requests: {} ## @section Persistence Parameters ## ## Enable persistence using Persistent Volume Claims ## ref: https://kubernetes.io/docs/concepts/storage/persistent-volumes/ ## persistence: ## @param persistence.enabled Enable persistence using Persistent Volume Claims ## enabled: true ## @param persistence.mountPath Path to mount the volume at. ## mountPath: /bitnami/pytorch ## @param persistence.subPath The subdirectory of the volume to mount to, useful in dev environments and one PV for multiple services ## subPath: "" ## @param persistence.storageClass Storage class of backing PVC ## If defined, storageClassName: ## If set to "-", storageClassName: "", which disables dynamic provisioning ## If undefined (the default) or set to null, no storageClassName spec is ## set, choosing the default provisioner. (gp2 on AWS, standard on ## GKE, AWS & OpenStack) ## storageClass: "" ## @param persistence.annotations Persistent Volume Claim annotations ## annotations: {} ## @param persistence.accessModes Persistent Volume Access Modes ## accessModes: - ReadWriteOnce ## @param persistence.size Size of data volume ## size: 8Gi ## @param persistence.existingClaim The name of an existing PVC to use for persistence ## existingClaim: "" ## @param persistence.selector Selector to match an existing Persistent Volume for WordPress data PVC ## If set, the PVC can't have a PV dynamically provisioned for it ## E.g. ## selector: ## matchLabels: ## app: my-app ## selector: {} ## @param persistence.dataSource Custom PVC data source ## dataSource: {}