- Setup infrastructure
Fill out terraform-setup/terraform.tfvars
with aws and digital ocean credentials.
make step_01
- Create 3 Node HA Cluster for Rancher
make step_02
- Install cert-manager and Rancher
make step_03
- Create 3 single-node downstream k3s clusters
make step_04
Go to https://rancher-demo.plgrnd.be/login and set up admin password and server url.
"Add Cluster" -> "Register an existing Kubernetes cluster" => "Other Cluster"
Add "group" label with values "amd" and "arm".
To register every cluster run
./run_on.sh [3-5] kubectl apply ....
Wait until all clusters are read.
Go to "Cluster Explorer" of local cluster -> "Continuous Delivery"
Add arm and amd Cluster Groups matching the cluster labels from above in "fleet-default".
Get Download Kubeconfig from all clusters.
Commands to watch clusters
watch kubectl --kubeconfig kubeconfig_cluster_one --insecure-skip-tls-verify get nodes,pods -A
watch kubectl --kubeconfig kubeconfig_cluster_two --insecure-skip-tls-verify get nodes,pods -A
watch kubectl --kubeconfig kubeconfig_cluster_three --insecure-skip-tls-verify get nodes,pods -A
Add Git repo to deploy system-upgrade-controller Repo: https://github.com/rancher/system-upgrade-controller Path: manifests All clusters.
Add Git Repo to deploy upgrade Plan Repo: https://github.com/bashofmann/rancher-k3s-fleet-examples Path: fleet-upgrades Only amd clusters
- Deploy upgrade plan
Add Git Repo to deploy rest Repo: https://github.com/bashofmann/rancher-k3s-fleet-examples Path: fleet-examples All clusters
- Add hello-world example
- Open webapp on nodeports
- Update hello-world example with overlays
- Show that on arm cluster the color changed
- Deploy netdata
- Change gitrepo cluster selector
To remove everything
make destroy