Kaptain DocumentationThis page lists Kaptain configuration properties that can be adjusted during the initial installation or by updating the existing instance of Kaptain. These parameters are typically set in params.yaml.
To learn more about overriding default parameter values as well as updating parameter on the existing Kaptain instance, refer to the Overriding parameters and the Update Parameters on Instance pages in KUDO documentation.
Kaptain Operator Parameters
| Parameter name | Description | Default value |
|---|---|---|
| backupBucket | Backup bucket name for the MySQL cluster | |
| backupEnabled | Enable backup for the MySQL cluster | False |
| backupEndpointUrl | Custom endpoint URL for the MySQL cluster backup bucket | |
| backupRegion | Backup bucket region name for the MySQL cluster | |
| backupRestoreSource | Full path to the backup folder, for example s3://mysql-store-backup/kaptain-mysql-store-2021-04-05-21:47:24-full. |
|
| backupSecretName | Name of the secret with access credentials for the MySQL cluster backup bucket | |
| base64CustomCertificate | base64-encoded contents of a custom certificate file (.crt) to use with the provided custom domain name | |
| base64CustomCertificateKey | base64-encoded contents of a custom certificate private key file (.key) to use with the provided custom domain name | |
| clusterDomainName | Kubernetes cluster domain name | cluster.local |
| customDomainName | Custom domain name used to access Kaptain, for example, kaptain.mycluster.company.com | |
| datastoreMySQLNodeCPU | CPU per single database node in the MySQL cluster | 2 |
| datastoreMySQLNodeDiskSize | Disk size of a single database node in the MySQL cluster | 30Gi |
| datastoreMySQLNodeMemory | Memory per single database node in the MySQL cluster | 2G |
| datastoreNodeCount | Number of MySQL nodes in the HA MySQL cluster | 3 |
| datastoreProxyCount | Number of proxy instances for the HA MySQL cluster | 2 |
| datastoreProxyNodeCPU | CPU per single proxy instance in the MySQL cluster | 1 |
| datastoreProxyNodeMemory | Memory per single proxy instance in the MySQL cluster | 1G |
| defaultNotebookImage | Default image to use when creating a new notebook server | mesosphere/kubeflow:1.3.0-jupyter-spark-3.0.0-horovod-0.22.0-tensorflow-2.5.0 |
| dkpPlatformVersion | The major version of the target DKP, used for Kaptain installation. | 2 |
| enableMultiUserSupport | Enable multi-user support for Pipelines | True |
| externalDexClientId | Dex client id to use when authenticating with external cluster | |
| externalDexClientSecret | Dex client secret to use when authenticating with external cluster | |
| installMinioOperator | Install Minio Operator to target namespace. Must be disabled in case there is another instance of the operator running in the cluster. | False |
| kubeflowIngressGatewayServiceAnnotations | Additional annotations for Kaptain’s default gateway | |
| minioServerPools | Number of MinIO server pools. | 1 |
| minioServers | The number of MinIO server pods to deploy in the pool. | 2 |
| minioStorageCapacity | The total capacity of a single MinIO server pool. | 40Gi |
| minioVolumeStorageClass | StorageClass name to be used for PVCs in MinIO server pool. | |
| minioVolumesPerServer | The number of Persistent Volume Claims to generate for each MinIO server pod in the pool. | 2 |
| notebookAffinityConfig | Pod affinity configurations for Notebook servers | |
| notebookCullingCheckPeriodMinutes | Notebook status check period | 1 |
| notebookEnableCulling | Enables scale down idling notebooks to freeing up the allocated resources. | False |
| notebookIdleTimeMinutes | The period of time a notebook can stay idle before it gets culled (scaled down) | 1440 |
| notebookImages | The list of available standard notebook Images | mesosphere/kubeflow:1.3.0-jupyter-spark-3.0.0-horovod-0.22.0-tensorflow-2.5.0 mesosphere/kubeflow:1.3.0-jupyter-spark-3.0.0-horovod-0.22.0-tensorflow-2.5.0-gpu mesosphere/kubeflow:1.3.0-jupyter-spark-3.0.0-horovod-0.22.0-pytorch-1.7.1 mesosphere/kubeflow:1.3.0-jupyter-spark-3.0.0-horovod-0.22.0-pytorch-1.7.1-gpu mesosphere/kubeflow:1.3.0-jupyter-spark-3.0.0-horovod-0.22.0-mxnet-1.9.0 mesosphere/kubeflow:1.3.0-jupyter-spark-3.0.0-horovod-0.22.0-mxnet-1.9.0-gpu |
| notebookTolerationGroups | Pod toleration group configurations for Notebook servers | |
| oidcProviderBase64CaBundle | CA bundle of the OIDC provider endpoint URL encoded in base64 | |
| oidcProviderEndpoint | External OIDC provider endpoint URL | |
| registrationFlow | Enable automatic profile creation | False |
| userIdHeader | Name of the HTTP header containing the information about authenticated user | kubeflow_user_id |
| userIdPrefix | Prefix for the ‘userIdHeader’ header value | kubeflow_user_id_prefix |
| workflowsTTLSecondsAfterFinish | The TTL for Argo Workflow (a single run of a pipeline) to persist after completion (default: 24 hours). | 86400 |
Automatic cleanup of completed and idling workloads
Automatic cleanup of completed and idling workloads…Read More
Configure Kaptain to authenticate with a Kommander Management Cluster
Configure Kaptain to authenticate with a Kommander Management Cluster…Read More
Configure a custom domain name and certificates for Kaptain
Configure a custom domain name and certificates for Kaptain…Read More
Configuring Notebook Servers
Configure Notebook Servers controls and settings…Read More