helm

Multi-Cluster Helm Example

This example will deploy the Kubernetes sample guestbook application as packaged as a Helm chart. The app will be deployed into the fleet-mc-helm-example namespace.

The application will be customized as follows per environment:

• Dev clusters: Only the redis leader is deployed and not the followers.
• Test clusters: Scale the front deployment to 3
• Prod clusters: Scale the front deployment to 3 and set the service type to LoadBalancer

kind: GitRepo
apiVersion: fleet.cattle.io/v1alpha1
metadata:
  name: helm
  namespace: fleet-default
spec:
  repo: https://github.com/rancher/fleet-examples
  paths:
  - multi-cluster/helm
  targets:
  - name: dev
    clusterSelector:
      matchLabels:
        env: dev

  - name: test
    clusterSelector:
      matchLabels:
        env: test

  - name: prod
    clusterSelector:
      matchLabels:
        env: prod

Step-by-Step

These are the steps I did to create a GitOps workflow

Forked fleet-examples

Just hit the "Fork" button above

Created GitRepo Resource in Rancher

• Select "Continuous Delivery" from the drop down left navigation menu
• Create a "Git Repo" resource using the form, making sure to set the repo desired path (multi-cluster/helm).
• Edited the YAML of "fleet-local" Clusters resource to add a env: prod label. This enabled the "prod" cluster selector in fleet.yaml