MinIO
MinIO is a Kubernetes native, high-performance object store with an S3-compatible API. The MinIO operator supports deploying MinIO tenants on any Kubernetes cluster.
To install the MinIO operator, follow these steps:
Add the MinIO repository to your local Helm configuration by running this Helm command:
helm repo add minio https://operator.min.io/ helm repo update
Install the MinIO operator in your Aizen installation by running this Helm command:
helm -n aizen-infra install aizen-minio-operator minio/operator --version 5.0.15
Copy the contents below to a file named
minio_values.yaml
, and change the storage class name fromstandard
to the storage class name that is defined in your Kubernetes cluster.Contents of minio_values.yaml:
enabled: false secrets: name: aizen-env-configuration accessKey: aizen secretKey: aizen@123 tenant: name: aizen configuration: name: aizen-env-configuration pools: - servers: 1 name: minio-pool volumesPerServer: 1 size: 50Gi storageClassName: standard mountPath: /opt/aizen subPath: /data certificate: requestAutoCert: false prometheusOperator: false logging: anonymous: true json: true quiet: true prometheus: disabled: true diskCapacityGB: 1 storageClassName: standard log: disabled: true db: volumeClaimTemplate: spec: storageClassName: standard accessModes: - ReadWriteOnce resources: requests: storage: 1Gi
Deploy the MinIO Tenant by running this Helm command:
helm -n aizen-infra install aizen-tenant --values minio_values.yaml minio/tenant --version 5.0.15
Check the deployment status of all the infrastructure components, including the MinIO operator, by running this command:
kubectl -n aizen-infra get pods
If any of the components are not in a Running state, see Installation Issues.
Run the MinIO Client (
mc
) as a container pod by running these commands:kubectl run my-mc -i --tty --image minio/mc:latest --rm --command -- bash mc --version mc alias set az http://minio.aizen-infra.svc.cluster.local aizen aizen@123 mc ls az
To create a bucket in MinIO for MLflow, run this command:
mc mb az/mlflow-artifacts exit
Last updated