Managed, patched control plane
HA control-plane pods on our shard, upgraded on a supported version ladder. You never babysit etcd or the API server.
Infrastructure Early access
We run and patch a highly-available control plane; you bring the nodes — homelab, Hetzner, AWS, a Pi in the closet — and join them as one cluster over an outbound-only WireGuard fabric.
— vector flow field · |v| across the plane
How it works
A managed, HA Kubernetes control plane runs on our shard. Your machines dial out to it over WireGuard — no inbound ports, no public IP — and register as nodes. You get one kubeconfig and standard tooling; nothing proprietary, nothing to unlearn.
What you get
HA control-plane pods on our shard, upgraded on a supported version ladder. You never babysit etcd or the API server.
Any Linux host that can dial out. Attach with one copy-paste command; detach whenever you like.
We charge for the control plane you use — never per worker node, never per vCPU you already own.
Upstream-conformant, your CNI, your workloads. One kubeconfig, and you can leave with it.
Pricing
You pay per control-plane replica. Each size is exactly double the last, so the price tracks the cost — and node counts are guidance, never a hard cap.
Small
€7 / replica / mo
A home cluster or a side project.
Medium
€14 / replica / mo
A small team running real workloads.
Large
€28 / replica / mo
Production across several clouds.
Redundancy is an orthogonal dial: add replicas for availability. Add-ons (backup & restore, public ingress, managed monitoring) attach per cluster and meter only where our cost scales — always with a cap you set and a €0 opt-out.
Part of the platform
Every Longbridge product shares one Org, one Console, and one bill — so adding the next one is a toggle, not another vendor to onboard.
See the whole platformTry it
# create a control plane
lbr cp create my-cluster --tier medium
# print a one-time join command, then run it on any Linux host
lbr cp nodes add my-cluster
# → curl -fsSL https://eu-hel.lngbrdg.com/join | sudo bash -s -- <token> The honest part
Start in the Console — or read the docs first. It's your call.