Early access — The platform is open but still being completed; features and prices may change before general availability.
FMJConsulting LogoFMJConsulting
Sign Up

Overview

What the FMJ Consulting IDP is, and the whole process in one picture.

The FMJ Consulting IDP is a control plane for your own infrastructure. You describe projects and pick components from a catalog; the platform provisions machines, releases charts, wires monitoring and keeps an audit trail — on your targets, driven by your CI runners. It never runs your workloads on our hardware, and it never holds your source code or application data.

The whole process, in one picture

Youpick componentsshape machinesconfirm by typingControl planeapp.fmjconsulting.frcatalogs (Ansible roles, Helm charts)projects, settings, plansaudit trailimmutable contracteverything the deploy will doYour CIGitLab pipelineruns on a runneryou bringTerraform · cloud-initAnsible · HelmYour infrastructureIncus — VMs, containersKubernetes — HelmAWS — EC2, EBSdeploycontract aspipelineappliesmetrics, health and deployment status flow back — dashboards show what the target really reportsYour credentials are used at the point of deployment, stored encrypted, scoped minimally. Your code and data never transit the platform.
You drive the control plane; it renders an immutable contract; a GitLab pipeline on your own runner applies it to your Incus, Kubernetes or AWS targets; metrics and status flow back to the dashboards.
  1. You attach components from the catalog to a project, shape the machines (image, CPU, memory, disk) or the releases (sizing), and confirm with a typed confirmation.
  2. The control plane renders everything into an immutable contract — the complete, reviewable description of what the deployment will do — and submits it as a GitLab pipeline. After your confirmation, changes apply automatically; nothing waits behind a hidden human step.
  3. Your CI executes the pipeline on a runner you bring, on your own machines. The runner applies the contract with Terraform, cloud-init, Ansible and Helm — your credentials are used there, at the point of deployment, and stored encrypted with minimal scopes.
  4. Your infrastructure receives the result: VMs and containers on Incus, Helm releases in your Kubernetes cluster, EC2 machines and EBS volumes in your AWS account, Nova machines and Cinder volumes in your OpenStack project. It belongs to your accounts — if the platform disappeared tomorrow, everything would keep running.
  5. Observability comes back: metrics collection and Grafana dashboards are provisioned with the components, and the health shown in the app is what the target actually reports — never an assumption.

Destroys travel the same road in reverse, as a cascade that removes everything the platform created, and a server-side sweeper finishes the job even if the browser goes away.

Where to go next

Early access

The platform is open but still being completed. Features and prices may change before general availability; the banner at the top of the app says so on every page while this is true.

Getting started

© 2026 FMJConsulting. All rights reserved.