Rafay Managed Kubernetes Clusters (Rafay MKS) - Rafay Product Documentation
Overview
This documentation provides an overview of system template for Rafay Managed Kubernetes Clusters. These templates are designed to simplify the provisioning, configuration, and management of Kubernetes clusters.
Initial Setup
The platform team is responsible for performing the initial configuration and setup of the MKS template. The sequence diagram below outlines the high-level steps. In this process, the platform team will configure and share the template from the system catalog to the project they manage and then share the template downstream with the end user.
End User ProjectSystem CatalogPlatform AdminEnd User ProjectSystem CatalogPlatform AdminSelects MKS Template from System CatalogShares Template with Predefined ControlsTemplate Available in End User's Project
End User Flow
The end user launches a shared template, provides required input values, and deploys the cluster.
Rafay Managed Kubernetes ClusterRafay ProjectEnd UserRafay Managed Kubernetes ClusterRafay ProjectEnd UserLaunches Shared Template for MKSProvides Required Input Values (API Key, Node Configuration, SSH Details)Clicks "Deploy"Provisions a Rafay Managed Kubernetes Cluster on the specified nodesCluster Deployed SuccessfullyProvides Kubeconfig File as Output
This system template allows you to configure, templatize, and provision a Rafay Managed Kubernetes Cluster (Rafay MKS) on any supported operating system. For more details, refer to this document.
The templates are designed to support both:
- Day 0 operations: Initial setup
- Day 2 operations: Ongoing management
Infrastructure Types
The template supports provisioning Rafay MKS clusters on various infrastructure types, including:
- Bare Metal: Users manage the lifecycle of hardware and the operating system.
- Virtual Machines: Supports Bring Your Own OS or pre-packaged images (e.g., QCOW2, OVA formats).
- Public Cloud: Flexible deployments on cloud infrastructure.
Key Capabilities
This template enables users to:
- Provision and manage the lifecycle of Rafay Managed Kubernetes Clusters.
- Configure:
- Container Network Interface (CNI)
- Add-ons defined in the cluster blueprint.
As part of the output, users receive a kubeconfig file with cluster-wide privileges for secure access.
Resources
This system template will deploy the following resources:
- Upstream Kubernetes on the specified nodes.
Pre-Requisites
API Key: Specify the API key of the controller for
API Keyinput variablePrivate SSH Key: Provide the SSH key of the node to run the installer for Kubernetes deployment.
Node Information:
- Specify details for:
- Control Plane Node(s)
- Worker Node(s)
- Specify details for:
Input Variables
Cluster Config
| Name | Default Value | Value Type | Description |
|---|---|---|---|
| System Components Placement | { "node_selector": {}, "tolerations": [] } |
JSON | Enter node selectors and tolerations for the cluster |
| High Availability (HA) | false |
Text | Allowed: [true, false] Select if HA should be enabled |
| Cluster Name | $(environment.name)$ |
Expressions | Enter the name of the Upstream Kubernetes cluster |
| Cluster Kubernetes Version | v1.31.4 |
Text | Allowed: [v1.32.0, v1.31.4, v1.30.8, v1.29.12] Select the Kubernetes version for the cluster |
| Network | { "cni":{"name":"Calico","version":"3.26.1"}, "pod_subnet":"10.244.0.0/16", "service_subnet":"10.96.0.0/12" } |
JSON | Enter the network information |
| Control Plane Node(s) | { "hostname-1": { "arch": "amd64", "hostname": "hostname-1", "private_ip": "10.1.0.67", "operating_system": "Ubuntu22.04", "roles": ["ControlPlane", "Worker"], "ssh": { "ip_address": "129.146.178.0", "port": "22", "private_key_path": "private-key", "username": "ubuntu" } } } |
JSON | Provide the control plane node information. The variable should match the node's hostname (e.g., 'hostname-1')` |
| Worker Node(s) | { "worker-1": { "arch": "amd64", "hostname": "worker-1", "private_ip": "10.1.0.68", "kubelet_extra_args": { "max-pods": "400", "cpu-manager-reconcile-period": "40s" } } } |
JSON | Provide the worker node information. The variable should match the node's hostname (e.g., 'worker-1') |
| Kubernetes Upgrade | { "strategy":"sequential", "params":{"worker_concurrency":"50%"} } |
JSON | Enter the upgrade strategy for the cluster |
| Cluster Labels | { "env": "dev", "release": "stable" } |
JSON | Enter any labels to assign to the cluster |
| Cluster Location | sanjose-us |
Text | Enter the location label where the cluster will be deployed |
| Cluster Project | $(environment.project.name)$ |
Expressions | Enter the project for the Upstream cluster |
| Auto Approve Nodes | true |
Text | Allowed: [true, false] Select if nodes should be auto-approved |
| Cluster Dedicated Control Planes | false |
Text | Allowed: [true, false] Select if dedicated control planes should be enabled |
| Kubelet Extra Args | { "max-pods": "300", "cpu-manager-reconcile-period": "30s" } |
JSON | Specify additional kubelet arguments if needed |
| Installer TTL | 300 |
Text | Time to live (TTL) for the installer pod, in seconds |
Blueprint Config
| Name | Default Value | Value Type | Description |
|---|---|---|---|
| Blueprint Name | default |
Text | Enter the name of the blueprint assigned to the cluster |
| Blueprint Version | latest |
Text | Specify the version of the blueprint for the cluster. For system blueprints, use "latest" |
Kata and OPA Config
| Name | Default Value | Value Type | Description |
|---|---|---|---|
| Enable Kata Deployment | false |
Text | Allowed: [true, false] Enable Kata containers support |
| Enable OPA Gatekeeper Deployment | false |
Text | Allowed: [true, false] Enable Open Policy Agent Gatekeeper integration |
| OPA Excluded Namespaces | `[\ |