Convert Docker Compose to Kubernetes once, never again

Compose gets you started, then the day you outgrow one box you rewrite everything for Kubernetes. Migrate your Compose services to standard manifests once instead, on a cluster that can grow. skube is a dedicated Kubernetes cluster we run for you on EU infrastructure, from €24/mo, flat.

See pricing Read the docs

From docker-compose.yml to kubectl apply

1 · Translate once

Turn your Compose services into Deployments and Services. kompose convert drafts them in seconds; cleaning them up takes an afternoon. One pass, from €24/mo.

2 · Ship on real k8s

Full cluster-admin on stock Kubernetes: any ingress, operator or CRD, exactly as it runs on the big clouds. 

kubectl apply -f ./manifests

3 · Grow, no rewrite

When uptime matters, apply the same manifests to an HA cluster, unchanged. No migration cliff, no lock-in.

Before: one Compose service

# docker-compose.yml
services:
  web:
    image: ghcr.io/you/app:1.2
    ports:
      - "80:8000"

After: a Deployment and a Service

# web.yaml
apiVersion: apps/v1
kind: Deployment
metadata: { name: web }
spec:
  selector: { matchLabels: { app: web } }
  template:
    metadata: { labels: { app: web } }
    spec:
      containers:
        - name: web
          image: ghcr.io/you/app:1.2
---
apiVersion: v1
kind: Service
metadata: { name: web }
spec:
  selector: { app: web }
  ports: [{ port: 80, targetPort: 8000 }]

Start from €24/mo

One flat price per cluster, billed monthly. A dedicated single-tenant cluster on its own server. Cancel anytime.

skube1

€24 /mo incl. VAT

2 vCPU · 4 GB · 40 GB SSD

A handful of small apps: side projects, staging, experiments.

  • ✓ Single-tenant cluster
  • ✓ Full cluster-admin
  • ✓ Ingress included
  • ✓ 20 TB traffic included
  • ✓ Nightly backups
  • ✓ Email support
  • ✓ Cancel anytime
Create a cluster
Recommended

skube2

€59 /mo incl. VAT

4 vCPU · 8 GB · 80 GB SSD

Early production: an app with real users, side projects alongside.

  • ✓ Single-tenant cluster
  • ✓ Full cluster-admin
  • ✓ Ingress included
  • ✓ 20 TB traffic included
  • ✓ Nightly backups
  • ✓ Email support
  • ✓ Cancel anytime
Create a cluster

skube3

€109 /mo incl. VAT

8 vCPU · 16 GB · 160 GB SSD

A busy cluster: many apps and bigger databases, side by side.

  • ✓ Single-tenant cluster
  • ✓ Full cluster-admin
  • ✓ Ingress included
  • ✓ 20 TB traffic included
  • ✓ Nightly backups
  • ✓ Email support
  • ✓ Cancel anytime
Create a cluster

Cancel anytime; EU consumers have a 14-day right of withdrawal.

Is one node enough?

For dev, staging and early production: yes. Each cluster runs on its own dedicated server, and even the smallest size (2 vCPU / 4 GB) comfortably runs a typical web app, its database and ingress for your first few thousand users. Bigger sizes go up to 8 vCPU / 16 GB. And skube runs on skube: the page you are reading is served from our own single-node cluster.

If the node fails, we restore your cluster from a nightly snapshot. And when you outgrow it, order the next size up and apply the same manifests, or take them to a multi-node HA cluster anywhere, unchanged. You pay for redundancy when you need it, not before.

Frequently asked

Can I move a Docker Compose app to Kubernetes here?
Yes, and that is the point. Instead of running Docker Compose until you hit the migration wall, you write Kubernetes manifests from day one on a small, cheap cluster. Translate your Compose services into Deployments and Services once, and you are on standard Kubernetes with nothing left to port later.
What happens to my Compose volumes?
They become PersistentVolumeClaims backed by your cluster’s SSD. A database from Compose moves over as a Deployment or StatefulSet with a PVC: the same data model, just declared.
Can I bring my own ingress or domain?
Yes. With full cluster-admin you can install any ingress controller, cert-manager and CRDs you like, and point your own domain at the cluster.
Do I have to rewrite my setup to scale later?
No. Because skube is stock Kubernetes, the manifests you apply here are the manifests you apply to an HA cluster when you outgrow one node. kubectl apply -f against the bigger cluster, unchanged. You pay the Kubernetes learning curve once, at hobby scale, and skip the rewrite entirely.
Who runs it?
We provision and stand up the cluster for you and keep the base healthy. You get full cluster-admin, so everything inside the cluster is yours to run.
Is my data backed up?
We keep nightly snapshots of your whole cluster (the last 7 days) as disaster recovery: if the server fails, we restore it. Treat them as a safety net, not an archive; a restore brings back the whole node as of last night, not single files. For anything you cannot afford to lose, also run your own backup tooling, you have full cluster-admin.

Move your Compose app to a cluster it can grow in

A dedicated cluster, a working kubeconfig, and manifests that follow you to HA.

Create a cluster

Free account; pay only when you order a cluster. Questions first? Email support@skube.dev, a human answers.