Skip to main content

Deploy a Kubernetes Cluster

Octos Cloud supports managed Kubernetes clusters with configurable master and worker nodes, auto-scaling, and integrated monitoring.

Cluster architecture

┌──────────────┐
│ Control │
│ Plane │
│ (Master) │
└──────┬───────┘

┌────────────┼────────────┐
│ │ │
┌─────┴─────┐ ┌───┴───┐ ┌──────┴────┐
│ Worker 1 │ │Worker 2│ │ Worker N │
│ (pods) │ │(pods) │ │ (pods) │
└───────────┘ └───────┘ └───────────┘

Create a cluster

  1. Navigate to Kubernetes in your project
  2. Click Create Kubernetes Cluster
  3. Configure:
ParameterDescription
Cluster NameDescriptive name for the cluster
Kubernetes VersionSelect the target version
ZoneData center region
Master nodesNumber and size of control plane nodes
Worker nodesNumber and size of worker nodes
SSH KeyFor node access
  1. Optionally enable auto-scaling with min/max worker counts
  2. Click Create

Access the cluster

Download kubeconfig

  1. Navigate to the cluster overview
  2. Click Download Kubeconfig
  3. Set the config:
export KUBECONFIG=~/downloads/kubeconfig-my-cluster.yaml
kubectl get nodes

Kubernetes Dashboard

# Run proxy locally
kubectl proxy

# Access dashboard at:
# http://localhost:8001/api/v1/namespaces/kubernetes-dashboard/services/https:kubernetes-dashboard:/proxy/

Generate a token for dashboard login:

kubectl -n kubernetes-dashboard create token admin-user

Auto-scaling

Configure auto-scaling policies to dynamically adjust worker node count based on resource utilization:

ParameterDescription
Min membersMinimum number of worker nodes
Max membersMaximum number of worker nodes
Scale up policyCPU/memory threshold to add nodes
Scale down policyCPU/memory threshold to remove nodes
CooldownWait time between scaling events

Next steps