Get started
How-to guides
Reference
Concepts
Self-host gallery
AppendixThe six questions
How does it reach my servers?
SSH and CLI shelling work, but they poll and parse text. Levelrail’s agent dials out over mutual TLS and uses the Docker Engine API, with no inbound port.
What happens when a deploy fails?
Traffic should cut over only after a real readiness probe passes, and the previous release should stay available. Levelrail pins prior images so cleanup cannot remove a rollback target.
Can I see what is happening?
Metrics, logs and alerts should not need a second stack. Levelrail keeps node-local metrics and log search in the core.
Where do secrets and data live?
Per-app keys wrapped by a master key held only by the control plane, and metrics and logs stay on your nodes.
What is the real cost?
Your servers plus your time. Levelrail is Apache 2.0 with one edition and no paywalled features.
How do I leave?
Importers and plain Docker underneath keep exit costs low. Levelrail can import from Coolify, Dokploy and CapRover read-only.
Where Levelrail sits
Positioning, not a ranking. Every project below is useful, and the full comparison page is candid about where each is strong.
Read the full comparison
Frequently asked questions
What is a self-hosted PaaS?
What is a self-hosted PaaS?
Software you run on your own servers that gives you a platform-as-a-service workflow: push code, get a running app with TLS, logs and rollback.
Who is Levelrail for?
Who is Levelrail for?
Operators running roughly 3 to 50 services on 1 to 10 Linux machines who do not want to run Kubernetes.
Is it a Kubernetes alternative?
Is it a Kubernetes alternative?
No. There is no bin-packing scheduler, autoscaling or CRD ecosystem. It borrows the reconciler pattern without the runtime.
How mature is it?
How mature is it?
Pre-release and moving fast. The case studies page lists what has been measured and what has not.