As per documentation install Rancher server on either a single node or HA mode.
The most basic installation is accomplished by running Rancher in Docker:
docker run -d --restart=unless-stopped \
-p 80:80 -p 443:443 \
--privileged \
rancher/rancher:v2.8-head
After running Rancher server, access Rancher server UI at https://<ip>
and create a new cluster; please refer to the documentation for more information about how to create a cluster, or follow the steps below:
- Upon accessing the UI, click
Add Cluster
, and select the Custom Cluster (Custom from existing nodes, AWS, Azure, DigitalOcean, Linode, vSphere). - For the purposes of this test, create the following 3-node nodegroup configuration:
- 1 node with all roles:
etcd, controlplane, worker
- 2 dedicated
worker
nodes
- 1 node with all roles:
- Ensure that k8s version
v1.27.5-rancher1-1
is selected - Disable the NGINX from the Advance Options
- Add the following extra_args: under the kube-api:
extra_args:
service-account-issuer: https://xx.xx.xx:6443 (xx.xx.xx.xx is controlplane public IP address)
extra_args_array:
service-account-issuer:
- rke
- Click
Create Cluster
- Make sure to add Public IP and Internal IP of node in Advance Options section.
- Run the registration command with the above mentioned configuration on already created nodes.
-
Once you Rancher Kubernetes cluster is active, Fetch it's kubeconfig.yml file and save it locally.
-
Download a sonobuoy binary release of the CLI, or build it yourself by running:
$ go install github.com/vmware-tanzu/sonobuoy@latest
- Configure your kubeconfig file by running:
$ export KUBECONFIG="/path/to/your/cluster/kubeconfig.yml"
- Run sonobuoy:
$ sonobuoy run --mode=certified-conformance
- Watch the logs:
$ sonobuoy logs
- Check the status:
$ sonobuoy status
- Once the status commands shows the run as completed, you can download the results tar.gz file:
$ sonobuoy retrieve
- To clean up Kubernetes objects created by Sonobuoy, run:
$ sonobuoy delete