Getting Started

Install

View as Markdown

Prerequisites

  • Kubernetes 1.28+ cluster with GPU nodes
  • kubectl configured and pointing at the target cluster
  • NVIDIA GPU Operator installed
  • Helm 3.x

Install the CLI

While this repository is internal, GitHub serves release assets only through authenticated API downloads — plain curl against releases/download/... returns a 404 “Not Found” page instead of the script, even with a token. Fetch the installer with the gh CLI (authenticate with gh auth login first). Set the version you want to install, then run the installer:

$export NVCRECTL_VERSION=v0.1.0-rc.9
$gh release download "${NVCRECTL_VERSION}" --repo NVIDIA/cluster-readiness-engine \
>  --pattern installer --output - | bash -s -- -v "${NVCRECTL_VERSION}"

The installer automatically downloads and verifies a SHA-256 checksum before installing. On air-gapped systems, ensure checksums.txt from the same release is reachable alongside the binary.

Verify the installation:

$nvcrectl --version

Set up the cluster

nvcrectl setup init installs the controller and its dependencies in two phases:

  1. deps — Kubeflow Trainer (required for TrainJob workloads)
  2. helm — CRE Helm chart (CRDs, controller deployment, built-in LogProfiles)
$nvcrectl setup init

GHCR authentication

The controller image and Helm chart are pulled from GHCR. Pass a GitHub token to authenticate — the CLI creates the pull secret for you:

$nvcrectl setup init --image-pull-secret $GITHUB_TOKEN

Verify

Check that the controller is running:

$kubectl get pods -n cluster-readiness-engine

Check that the CRDs are installed:

$kubectl get crds | grep cre.nvidia.com

Uninstall

$nvcrectl setup reset

This removes all CRE custom resources, the controller, CRDs, and Kubeflow Trainer. To keep Kubeflow Trainer, pass --skip-phases=deps:

$nvcrectl setup reset --skip-phases=deps