# README
Cluster-Forge
A helper tool that sets up all essential platform tools to prepare a Kubernetes cluster for running applications.
Overview
Cluster-Forge is a tool designed to bundle various third-party, community, and in-house components into a single, streamlined stack that can be deployed in Kubernetes clusters. By automating the process, Cluster-Forge simplifies the repeated creation of consistent, ready-to-use clusters.
This tool is not meant to replace simple helm install or kubectl apply commands for single-use development clusters. Instead, it wraps these workflows into a robust process tailored for scenarios such as:
- Ephemeral test clusters
- CI/CD pipeline clusters
- Scaling and managing multiple clusters efficiently
Cluster-Forge is built with the idea of ephemeral and reproducible clusters, enabling you to spin up identical environments quickly and reliably.
Usage
To deploy a ClusterForge SW stack, download a release package, and run 'deploy.sh'. This assumes there is a working kubernetes cluster to deploy into, and the current KUBECONFIG context refers to that cluster.
While ClusterForge does not in any way require AMD Instinct GPU's, this was a primary use case during intial development. For ease of use in such a server, a helper script is available to deploy RKE2 kubernetes, install rocm, and then setup ClusterForge. It can be found in the setup.sh file. It can also be run with:
curl https://raw.githubusercontent.com/silogen/cluster-forge/refs/heads/main/setup.sh | sudo bash
Storage Classes
Storage classes are provided by default with Longhorn. These can be changed or customized as needed.
Out of the box, these are the storage classes and purposes:
| Purpose | storageclass | Type | Locality | |
|---|---|---|---|---|
| GPU Job | mlstorage | RWO | LOCAL/remote | |
| GPU Job | default | RWO | LOCAL/remote | |
| Ask / know what you're doing | direct | RWO | LOCAL | |
| Multi-container | multinode | RWX | ANYWHERE |
Known Issues
Cluster-Forge is still a work in progress, and the following issues are currently known:
- Terminal Line Handling: Errors occurring alongside the progress spinner may cause terminal formatting issues. To restore the terminal, run:
reset
Future Improvements
We are actively working on resolving the known issues and improving the overall functionality of Cluster-Forge. Your feedback is always welcome!
Conclusion
Cluster-Forge is designed to simplify Kubernetes cluster management, especially when dealing with ephemeral, test, or pipeline clusters. By combining multiple tools and workflows into a repeatable process, Cluster-Forge saves time and ensures consistency across deployments.
Give it a try, and let us know how it works for you!