...
- From the Morpheus portal, access the “Provisioning”→”Instances” , access the "Rancher" instance page and go under "Actions"→"Run workflow" :
- Search for the workflow "Deploy RKE2 cluster" :
- Fill in the necessary parameters:
- Rancher Dashboard Admin Password : this the password that was set to access the Rancher Dashboard instance
- RKE2 Cluster Name: name for the cluster that will appear on the dashboard
- RKE2 Cluster Environment: tag for the cluster (dev, stage, prod)
- RKE2 Cluster Kubernetes version: version of the kubernetes cluster (e.g. 1.28)
- RKE2 Cluster worker node plan: resource plan for the worker node (e.g. 8cpu-8gbmem-30gbdisk)
- RKE2 Cluster Kubernetes deployment mode: "Development" (1 controller node) or "High Availability" (3 controller nodes)
- RKE2 Cluster Autoscaler maximum number of nodes: upper limit for scaling up the size of the cluster (minimum is set to 2 worker nodes)
- Leave the "Command Options" field empty
Once filled the parameters, press "Execute" :
- Monitor the execution of the workflow via the progress bar and the "History" tab for more information:
- Wait for the workflow to finish. Once completed, the new cluster should be connected and visible in the Rancher Dashboard:
- If needed, you can provision and connect multiple clusters to the Rancher Dashboard. You just need to re-execute the workflow to deploy the additional clusters.
| Warning | ||
|---|---|---|
| ||
Please note that the provisioned clusters are equipped with an autoscaler component that is able to scale up or down the worker nodes depending on the load. Currently it is acknowledged a known issue which consists on the lack of cleanup of older ssh-keys after the worker nodes deletion by the autoscaler, which remains in the underlying Openstack project. Therefore after a frequent scale up/down it is possible to hit the limits of the Openstack ssh-keys quota preventing the creation of additional nodes by the autoscaler. In case you hit this problem please raise an issue in the EWC Support Portal. |