Skip to main content

Adding a On-premise Cluster

This section provides a step-by-step guide to adding and onboarding an on-premises Kubernetes cluster within the Nexastack platform.
Following these steps ensures that your cluster is configured correctly and ready for AI model deployment.

Step 1: Initiate Cluster Creation​

  1. Navigate to the Clusters section from the sidebar or dashboard.
  2. Click Add New Cluster.

Add New Cluster Button

Step 2: Select Cluster Type​

  1. Choose On-Premises as the cluster type.
  2. A configuration popup will appear prompting you to enter the cluster details.

Cluster Type Selection

Cluster Types

Nexastack supports multiple Kubernetes cluster types:

  • Cloud Clusters: Hosted and managed by cloud providers for easy scalability.
  • On-Premises Clusters: Deployed within your organization’s infrastructure for full control and compliance.
  • Managed by Nexastack: Fully handled by Nexastack, including provisioning, updates, and monitoring.

Choose the cluster type that best aligns with your operational and infrastructure needs.

Step 3: Configure Cluster Details​

Fill in the required fields in the configuration popup:

FieldExample Value
Cluster NameTeamX-Dev-Cluster
Prometheus URLhttps://prometheus.example.com
Ingress URLhttps://ingress.example.com

Once all details are entered, click Proceed Next.

Cluster Configuration Popup

Naming & Configuration Guidelines
  • Use clear, descriptive names to easily identify your clusters (e.g., akira-ai-onprem or ml-prod-cluster).
  • Validate that Prometheus and Ingress URLs are correct and reachable within your network.
  • Ensure the Ingress URL points to the appropriate domain or subdomain used by your Kubernetes services.

Step 4: Upload the Kubeconfig File​

The kubeconfig file is essential for Nexastack to connect and manage your Kubernetes cluster.

  1. Download the kubeconfig file from your cluster’s control plane and save it as a .yaml file.

    Secure Handling

    The kubeconfig file contains sensitive credentials. Store it securely and avoid sharing it externally.

  2. Upload the downloaded kubeconfig file in the Nexastack interface.

  3. Click Proceed Next to continue.

Upload Kubeconfig

Access Permissions

If the uploaded kubeconfig does not have adequate permissions, onboarding may fail.
Ensure the user or service account in the kubeconfig file has admin or cluster-wide privileges.

Step 5: Confirmation​

  1. A popup confirmation will appear once the kubeconfig file is uploaded successfully.

Success Message

  1. The newly onboarded cluster will now appear on the Clusters page, displaying its name, type, and status.

Clusters Page

Best Practices​

  • ✅ Verify Access: Confirm kubeconfig has appropriate permissions before onboarding.
  • 🏷️ Use Descriptive Names: Helps in managing multiple clusters efficiently.
  • 🌐 Validate Ingress: Ensure your ingress domain and class configuration are accurate.
  • 🔒 Keep Kubeconfig Secure: Treat it like a password; restrict access to authorized users only.

Next Steps​

  • Deploy your first AI model using the Model Deployment Guide.
  • Monitor cluster performance and resource utilization through the Clusters Dashboard.
  • Repeat this process to onboard additional clusters as your infrastructure grows.
Cluster Onboarded Successfully

Congratulations! 🎉
Your Kubernetes cluster has been successfully onboarded to Nexastack and is now ready for deploying and managing AI workloads.