Kubernetes
Managed container orchestration with virtual cluster abstraction
Managed Kubernetes clusters on Tasmanian infrastructure. Deploy and manage containerized workloads with our virtual cluster abstraction—no need to manage underlying infrastructure.
Overview
Our Kubernetes offering provides:
- Virtual cluster architecture — Full Kubernetes control plane without dedicated hardware
- Multi-tenant isolation — Secure separation between workloads
- FluxCD GitOps — Declarative cluster management
- VPN-only access — Secure cluster access via Netbird
- Sovereign infrastructure — 100% Tasmanian data residency
What We're Solving
Traditional Kubernetes requires significant operational overhead:
- Control plane management — etcd, API server, controllers need expertise
- Infrastructure provisioning — Bare metal or VM orchestration
- Upgrade coordination — Complex rolling update procedures
- High availability — Multi-node control plane for production
Our virtual cluster abstraction handles all of this for you.
Virtual Cluster Architecture
Instead of provisioning dedicated hardware, each cluster runs as a virtual cluster—a fully functional Kubernetes control plane that shares underlying infrastructure with other virtual clusters.
How It Works
graph TB
subgraph "Shared Infrastructure"
subgraph "Host Cluster"
HC[Host Kubernetes]
subgraph "Virtual Cluster 1"
VC1_API[API Server]
VC1_ETCD[etcd]
VC1 Controller[Controllers]
end
subgraph "Virtual Cluster 2"
VC2_API[API Server]
VC2_ETCD[etcd]
VC2 Controller[Controllers]
end
subgraph "Virtual Cluster 3"
VC3_API[API Server]
VC3_ETCD[etcd]
VC3 Controller[Controllers]
end
Workers[Worker Nodes<br/>Shared Pool]
end
end
User1[Your Team] --> VC1_API
User2[Other Team] --> VC2_API
User3[Other Team] --> VC3_API
VC1_API --> Workers
VC2_API --> Workers
VC3_API --> Workers
Benefits
| Aspect | Traditional K8s | Virtual Clusters |
|---|---|---|
| Provisioning | Hours to days | Minutes |
| Control plane cost | Dedicated nodes | Shared, efficient |
| Isolation | Separate clusters | Strong multi-tenant |
| Management | You manage | We handle upgrades |
| Scaling | Add physical nodes | Instant allocation |
What You Get
Each virtual cluster includes:
- ✅ Dedicated Kubernetes API server
- ✅ Isolated etcd datastore
- ✅ Full RBAC permissions
- ✅ Custom resource definitions
- ✅ Namespaced resource quotas
- ✅ Independent upgrades
What We Handle
We manage:
- 🔧 Control plane high availability
- 🔧 etcd backups and recovery
- 🔧 Kubernetes version upgrades
- 🔧 Security patches
- 🔧 Certificate rotation
- 🔧 API server scaling
Use Cases
- Microservices applications — Container orchestration at scale
- Development teams — Isolated environments per team/project
- CI/CD pipelines — Ephemeral clusters for testing
- SaaS platforms — Multi-tenant application hosting
What We Don't Offer
- ❌ GPU nodes (no GPU hardware available)
- ❌ Public load balancers with public IPs
- ❌ Direct host cluster access
Getting Started
1. Create a Cluster
Via O2S Portal:
- Go to o2s.tasmanian.cloud
- Navigate to Kubernetes > Create Cluster
- Select configuration (size, Kubernetes version)
- Deploy
Via CLI:
# Create a cluster
tc k8s create \
--name production \
--version 1.29
# Check status
tc k8s status production
# List clusters
tc k8s list
2. Configure kubectl
# Download kubeconfig
tc k8s kubeconfig production > ~/.kube/config
# Verify connection
kubectl get nodes
3. Connect via VPN
Cluster API is only accessible via Netbird VPN:
# Join your network
netbird up --setup-key YOUR_CLUSTER_KEY
# Verify connectivity
kubectl cluster-info
Cluster Sizing
| Plan | Control Plane | Best For |
|---|---|---|
| Starter | Shared | Development, testing |
| Standard | Dedicated | Production workloads |
| Pro | Dedicated + HA | Critical applications |
Available Versions
We support current Kubernetes releases:
- 1.28
- 1.29
- 1.30
Storage
Default Storage Class
Clusters include distributed block storage:
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
name: my-data
spec:
accessModes:
- ReadWriteOnce
storageClassName: default
resources:
requests:
storage: 10Gi
Object Storage
For object storage, use our S3-compatible service:
apiVersion: v1
kind: Secret
metadata:
name: s3-credentials
type: Opaque
stringData:
access-key: YOUR_ACCESS_KEY
secret-key: YOUR_SECRET_KEY
Networking
Network Policies
Secure your workloads with network policies:
apiVersion: networking.k8s.io/v1
kind: NetworkPolicy
metadata:
name: api-allow-frontend
spec:
podSelector:
matchLabels:
app: api
policyTypes:
- Ingress
ingress:
- from:
- podSelector:
matchLabels:
app: frontend
ports:
- protocol: TCP
port: 8080
Service Types
| Type | Description |
|---|---|
ClusterIP | Internal cluster access only |
NodePort | Expose on node IP (VPN accessible) |
LoadBalancer | Creates internal load balancer |
GitOps with FluxCD
All clusters come with FluxCD pre-installed.
Bootstrap Flux
flux bootstrap github \
--owner=myorg \
--repository=k8s-config \
--branch=main \
--path=clusters/production
Automated Deployments
Push to your repo, Flux automatically applies changes:
git add .
git commit -m "Deploy v1.2.3"
git push
Monitoring
Every cluster includes:
- Prometheus — Metrics collection
- Grafana — Dashboards
- Loki — Log aggregation
Access Dashboards
kubectl port-forward -n monitoring svc/grafana 3000:3000
# Access via VPN at http://localhost:3000
Security
RBAC
Full RBAC support with user-based access:
apiVersion: rbac.authorization.k8s.io/v1
kind: Role
metadata:
name: developer
rules:
- apiGroups: ["", "apps"]
resources", "deployments: ["pods", "services"]
verbs: ["get", "list", "watch", "create", "update", "patch"]
Pod Security
Clusters enforce restricted pod security standards by default.
Upgrades
We handle control plane upgrades automatically:
# Check current version
kubectl version
# Request upgrade
tc k8s upgrade production --version 1.30
Pricing
| Component | Price |
|---|---|
| Starter cluster | $25/month |
| Standard cluster | $50/month |
| Pro cluster | $100/month |
| Storage | $0.10/GB/month |
Managed (Dedicated Hardware)
For organizations requiring dedicated infrastructure, we offer Managed Kubernetes on dedicated hardware.
What's Included
- Dedicated control plane nodes
- Dedicated worker nodes
- Custom networking configuration
- Direct hardware access
- Custom storage options
Contact Us
This offering requires custom configuration. Contact us to discuss your requirements:
Related Documentation
- VPS — Single virtual machine workloads
- Templates — One-click application deployments
- Storage — Object and block storage
- API Reference — REST API