Kubernetes

The Control Panel provides multi-cluster Kubernetes management. You can register multiple Kubernetes clusters, store their kubeconfigs with AES-256 encryption, monitor cluster health, and deploy workloads across clusters from a single interface.

Cluster Registration

Adding a Cluster

1. Navigate to Kubernetes from the sidebar
2. Click Register Cluster
3. Enter a display name for the cluster
4. Paste the kubeconfig contents or upload a kubeconfig file
5. Click Register

The Control Panel encrypts the kubeconfig with AES-256 using the CC_K8S_ENCRYPTION_KEY environment variable before storing it in the database. The kubeconfig is only decrypted in memory when the Control Panel needs to communicate with the cluster.

Kubeconfig Security

Health Checking

The Control Panel performs health checks against registered clusters at a 2-minute interval. Each health check verifies:

• API server reachability — can the Control Panel connect to the cluster’s API server
• Node status — are cluster nodes in a Ready state
• System pod health — are critical system pods (kube-system namespace) running

Cluster Statuses

Health status is displayed on the Kubernetes page and updates automatically as checks run.

Unified Cluster View

The Kubernetes page provides a consolidated view of all registered clusters:

• Cluster cards — each cluster shown with its name, status, node count, and last health check time
• Resource summary — aggregated CPU and memory capacity across all clusters
• Namespace list — namespaces per cluster with workload counts

Click any cluster card to see its detailed view including nodes, namespaces, deployments, and pods.

Cross-Cluster Deployments

The Control Panel can deploy workloads to one or more clusters from a single operation.

Deployment Kinds

The following Kubernetes resource types are supported for cross-cluster deployments:

Creating a Cross-Cluster Deployment

1. Navigate to Kubernetes > Deployments
2. Click New Deployment
3. Select the deployment kind
4. Configure the workload (image, replicas, environment variables, volumes)
5. Select the target clusters — one or more registered clusters
6. Click Deploy

The Control Panel applies the resource definition to each selected cluster and tracks the deployment status per cluster.

Deployment Status

Each deployment shows per-cluster status:

• Deploying — resource is being applied to the cluster
• Running — all pods are in a Ready state
• Degraded — some pods are not ready
• Failed — deployment could not be applied

Cluster Detail

The cluster detail page provides a deeper look at a single cluster:

Nodes

• Node name, status, role (control-plane, worker)
• CPU and memory capacity and allocation
• Pod count per node
• Conditions and taints

Namespaces

• List of namespaces with workload counts
• Resource quotas if configured
• Click a namespace to see its workloads

Workloads

• Deployments, StatefulSets, DaemonSets, and Jobs
• Replica status (desired vs. ready)
• Pod list with status, restarts, and age

Pods

• Pod name, namespace, node, and status
• Container images and resource usage
• Logs accessible from the pod detail view

Managing Clusters

Editing a Cluster

Update a cluster’s display name or replace its kubeconfig from the cluster detail page. Click Edit and update the desired fields.

Removing a Cluster

Remove a cluster registration from the Control Panel. This does not affect the cluster itself — it only removes the Control Panel’s reference to it.

1. Open the cluster detail page
2. Click Remove Cluster
3. Confirm the removal

Next Steps

• High Availability — set up Control Panel redundancy
• Fleet Management — manage Cloud OS instances alongside K8s clusters