Skip to main content

Lumi Enterprise on OpenShift overview

info

This information applies to Imply Lumi Enterprise deployments on OpenShift.
Lumi Enterprise on OpenShift is an early access feature.

You can deploy Imply Lumi Enterprise on OpenShift, the Red Hat enterprise distribution of Kubernetes. Use this deployment method if you want to adopt Lumi Enterprise into your existing infrastructure or run Lumi on bare metal.

Lumi Enterprise provides platform-agnostic Helm charts that provision and deploy Lumi services into a Kubernetes environment. The charts connect Lumi to the external dependencies that you provide.

The topics in this section describe how to deploy Lumi Enterprise on bare-metal OpenShift, where you provision the infrastructure yourself. They cover how to prepare your environment, install Lumi Enterprise, and verify your deployment.

Requirements​

Before you install the Lumi Helm chart on a bare-metal OpenShift cluster, make sure you have the following:

ComponentVersionUsed for
OpenShift cluster4.19 or newer (Kubernetes 1.32+)Runs all Lumi workloads
Helm3.18 or newerInstalls Lumi
MySQL databaseMySQL 8.0 or newerMetadata store for all Lumi services
Kafka cluster3.8 or newerEvent ingestion pipeline
S3-compatible object store—Deep storage
Load balancer, DNS, and TLS—External access to the Lumi gateway

You don't need a dedicated OpenShift cluster to deploy Lumi. Lumi can run alongside other applications. For more information on supported Kubernetes versions with OpenShift deployments, see Red Hat OpenShift Container Platform Life Cycle Policy.

Deploy Lumi Enterprise​

To deploy Lumi Enterprise on bare-metal OpenShift, complete the following topics in order:

  1. Obtain credentials to get your Docker Hub credentials, Lumi onboarding credentials, and RSA key pair.
  2. Configure permissions to grant Lumi the service account and cluster access it needs.
  3. Provision dependencies to set up your database, Kafka cluster, and object store.
  4. Set up networking and ingress to configure the load balancer, DNS, TLS, and ingress controller.
  5. Size and prepare nodes to plan node sizes, local disks, and kernel settings.
  6. Configure node labels and taints to control which OpenShift nodes run each Lumi node type.
  7. Install Lumi Enterprise using the Helm chart, and learn how to upgrade and uninstall your deployment.
  8. Verify your deployment to confirm that your deployment is healthy.

We suggest applying the following configurations to your environment:

  • Local NVMe storage: For best segment storage performance, use local NVMe storage for data nodes. See Local disks for data nodes.
  • Autoscaling: Enable node autoscaling so upgrades and ingestion don't stall. See Autoscaling.
  • Database high availability: Run your database with a primary and at least one replica. See MySQL database.
  • Kafka broker spread: Spread Kafka brokers across separate failure domains. See Kafka cluster.

Learn more​

See the following topics for more information: