Cluster Management
Overview
Cloud Orbiter provides comprehensive Kubernetes cluster lifecycle management — from initial provisioning to ongoing Day 2 operations — across any infrastructure type. It aims to provide an excellent user experience in managing hundreds of thousands of clusters and maintaining large-scale distributed applications.
Core capabilities:
- Deploy, create, manage, monitor, and upgrade multiple clusters across geo-distributed environments
- Manage all Kubernetes resources: nodes, events, namespaces, workloads, pods, ReplicaSets, Deployments, DaemonSets, StatefulSets, access control, network policies, storage, secrets, and ConfigMaps
- Connect multiple clusters so that all management flows through the Cloud Orbiter control plane
- Remote cluster management via outbound agent — no inbound firewall rules required
Kubernetes Engine
Cloud Orbiter's Kubernetes engine is CKP — Coredge's enterprise-grade, CNCF-certified Kubernetes distribution, purpose-built for on-premise and bare metal. It delivers production clusters with zero distribution-management work on your part.
For the full version stack, provisioning internals, engine components, and built-in add-ons, see Kubernetes Engine.
Creating a Cluster
- Navigate to Clusters from the left navigation panel
- Click + Add Cluster
- Select Create Cluster
- Choose cluster type and provide:
- Cluster name and description
- Kubernetes distribution and version
- Networking and networking version
- Provider configuration
- Master nodes: count, host group, virtual IP
- Worker nodes: count, host group
- Click Create — a real-time cluster state dashboard tracks provisioning progress
Day 2 Operations
Cloud Orbiter automates the ongoing management tasks that keep clusters healthy and efficient:
Cluster Upgrades
Cloud Orbiter automates upgrading Kubernetes clusters to the latest version — ensuring clusters run on the latest stable release with up-to-date security patches and bug fixes. Upgrades are performed without application downtime.
Scaling
Add or remove cluster nodes through a centralized interface. Cloud Orbiter handles the node provisioning, Kubernetes registration, and workload redistribution automatically.
Monitoring
Real-time monitoring tracks:
- Node health status and availability
- CPU and memory usage per node and per workload
- Network I/O pressure
- Pod capacity vs. usage
Security
Day 2 security operations include:
- RBAC enforcement on every API call
- Network policies managing traffic between pods and namespaces
- Container image scanning for vulnerability detection
Backup and Restore
Cluster data protection through Velero integration — see Backup & Restore for details.
Cluster Monitoring
Health Monitoring
View the health status of each node and receive alerts when nodes become unhealthy. Quickly identify issues impacting cluster stability.
CPU and Memory Monitoring
Real-time CPU and memory usage tracking per node and per workload. Identify resource bottlenecks and performance issues as they occur.
Node Monitoring
Per-node metrics including:
- CPU utilization vs. capacity
- Memory usage vs. capacity
- Disk usage
- Network activity
Alerts can be configured to notify administrators when thresholds are exceeded.
Cluster Access Management
Service Accounts
Map Cloud Orbiter users to Kubernetes service accounts for fine-grained RBAC on target clusters:
- Click Map User in the Cloud Orbiter Management Console
- Enter the Namespace and Service Account for the mapping
Users can only perform actions on the target cluster that are permitted by their mapped service account.
Cluster Terminal
Access the Kubernetes CLI directly from the Cloud Orbiter dashboard:
- Navigate to the cluster dashboard
- Select the connected cluster
- Click the Terminal button (top-right corner)
- Execute kubectl commands directly — pods, namespaces, configmaps, secrets, and more
Kubeconfig Download
Download the kubeconfig file to access the cluster from a local machine:
- Log in and navigate to the cluster
- Click Download Kubeconfig from the top navigation menu
- Save the file and use with kubectl locally
Location Tagging
Cloud Orbiter supports geo-aware cluster management with location tagging:
- Automatically detect cluster geolocation based on public IP
- View clusters in List View (status, version, distribution, health, pods, utilization) or Map View (graphical world map representation)
- Click a cluster marker on the map to view details and access cluster management
Service Discovery
Services
Cloud Orbiter exposes Kubernetes services for application access. Services provide network discovery without modifying application code — a Service makes the set of Pods running an application available to clients.
Centralized Ingress
Cloud Orbiter's Centralized Ingress enables cross-cluster service access without:
- Writing Ingress configuration files
- Creating NodePorts or Load Balancers manually
- External tenant or IP-based configuration
Applications are accessible across clusters automatically once services are created — Cloud Orbiter handles the routing.
Test Suites
Cloud Orbiter includes pre-configured test suites with pre-packaged test cases to:
- Verify cluster health before application deployment
- Confirm cluster performance after deployment
- Run proactive checks as part of ongoing operations
Test suite execution results are downloadable for audit and reporting purposes.