Install and Use Fleet CLI

SUSE® Rancher Prime Continuous Delivery CLI is a command-line interface (CLI) that allows you to interact directly with SUSE® Rancher Prime Continuous Delivery from your local machine. It enables you to create, apply, and inspect bundles without requiring a GitRepo. Typical use cases include:

  • Testing and previewing bundle contents.

  • Creating bundles directly from Helm charts, Kubernetes manifests, and fleet.yaml files.

  • Checking which clusters a bundle would target.

  • Validating deployments without installing SUSE® Rancher Prime Continuous Delivery in the cluster.

You can use fleet apply without installing SUSE® Rancher Prime Continuous Delivery in your clusters. However, for cluster interaction (for example, fleet target, fleet deploy), SUSE® Rancher Prime Continuous Delivery must be installed.

For more information, see Install Fleet.

Install Fleet CLI

Fleet CLI is a stand-alone binary that you can download from the Fleet GitHub releases page.

Linux/macOS

curl -L -o fleet https://github.com/rancher/fleet/releases/latest/download/fleet-linux-amd64

# Make it executable and move to PATH
chmod +x fleet
sudo mv fleet /usr/local/bin/

Windows (PowerShell)

Invoke-WebRequest -Uri "https://github.com/rancher/fleet/releases/latest/download/fleet-windows-amd64.exe" -OutFile "fleet.exe"

Verify installation

fleet --version

Prerequisites

Make sure you have the following tools installed and configured:

  • A working Kubernetes cluster (for example, k3s, kind, or a cloud provider).

  • kubectl is configured for your cluster.

  • Helm is installed.

  • Fleet CLI is installed and accessible in your terminal.

Verify prerequisites

kubectl get nodes
helm version
fleet --version

Key Commands

SUSE® Rancher Prime Continuous Delivery provides several CLI commands to create, preview, and deploy bundles. These commands are useful for debugging and understanding the bundle lifecycle.

  • fleet apply: Creates or previews a bundle from local files, such as a Helm chart, Kubernetes manifests, or kustomize folders. This command does not require access to a cluster, so it works even without SUSE® Rancher Prime Continuous Delivery or kubectl installed.

    Example:

fleet apply my-bundle ./manifests

  • fleet target: Reads a bundle file and evaluates which clusters would receive it, based on selectors and targeting rules such as targets, targetOverrides, clusterGroups, and label selectors.

    Example:

fleet target my-bundle ./manifests

  • fleet deploy: Takes the output of fleet target, or a dumped bundledeployment/content resource, and deploys it to a cluster—just like fleet-agent would.

    You can use it in these scenarios: fleet apply -o - name ./folder to check the YAML of the bundle before creating it. For more information, see Examining the Bundle Lifecycle with the CLI. Use with a target to debug selectors and verify which downstream clusters are targeted. ** fleet deploy --dry-run to print resources that would be deployed, without applying them.

A diagram explaining how Fleet CLI key commands work

Deploy a Sample Bundle Using Fleet CLI

You can deploy workloads without using GitRepos by applying them locally with the CLI. For example, use the Fleet examples repository:

git clone https://github.com/rancher/fleet-examples
cd fleet-examples/single-cluster

Apply it to the current cluster:

fleet apply -o my-cool-bundle manifests

This command creates a Bundle resource in the namespace.

Convert a Helm Chart into a Bundle

You can use Fleet CLI to convert a Helm chart into a bundle. For example, to convert the external-secrets operator chart:

cat > targets.yaml <<EOF
targets:
- clusterSelector: {}
EOF

mkdir app
cat > app/fleet.yaml <<EOF
defaultNamespace: external-secrets
helm:
  repo: https://charts.external-secrets.io
  chart: external-secrets
EOF

fleet apply --compress --targets-file=targets.yaml -n fleet-default -o - external-secrets app > eso-bundle.yaml

kubectl apply -f eso-bundle.yaml

Make sure you use a cluster selector in targets.yaml that matches all clusters you want to deploy to.

For more information, refer to Fleet: Multi-Cluster Deployment with the Help of External Secrets.

Troubleshooting

If the bundle is not ready:

  • Check if fleet-controller and fleet-agent pods are running.

  • Make sure the fleet-local cluster is registered.

  • Inspect the bundle for error messages:

    kubectl describe bundle -n fleet-local <bundle-name>

  • Delete and reapply the bundle if you encounter Helm ownership conflicts.

Verify Agent and Controller Status

Before troubleshooting bundle or deployment issues, verify that the Fleet agent is registered and running on the downstream cluster. For more information, see Fleet Agent is Registered, Watches for BundleDeployments.

If a bundle is not ready on a given cluster, check the following:

  • On the management cluster:

    • Verify that the fleet-controller and gitjob pods (or helmops, depending on your setup) are running.

    • Ensure the cluster status shows Ready.

    • Check the status of the bundle—it should contain an error message if deployment failed.

  • On the target cluster where the bundle isn’t ready:

    • Verify that this target cluster is registered and has a running fleet-agent pod.

    • As a last resort, check the logs of the fleet-agent pod.