SUSE Virtualization: Hands-on Workshop
The Virtualization Rodeo
Vertex Trust Bank • Harvester HCI • Rancher Prime • deployed with rodeo-cli
Choose your platform →
Start the lab →
Quick reference
The scenario¶
Vertex Trust Bank is drowning in legacy hypervisor renewal costs. Sarah, the CTO, hands you the keys to a fresh SUSE Virtualization platform and a list of problems that will not wait. Across eight exercises you bring the cluster online, ship a VM under pressure, dodge a hardware failure with live migration, wall off a sensitive network, recover a deleted record, and scale a fleet from templates.
This workshop is the self-hosted twin of SUSE's customer-facing
Virtualization Rodeo. Same story,
same skills, same 8 exercises regardless of where you deploy it. Only the
infrastructure bring-up step differs per platform, with its own precise
instructions below, driven by
rodeo-cli instead of joining Instruqt.
Choose your platform¶
Bare metal
Your own KVM host
⏱ 90-150 min
Min: 64 GiB RAM / 32 vCPU / ~1 TB disk. Recommended: 80-96 GiB / 40+ vCPU. /dev/kvm present, survives SSH drops via tmux.
AWS
Fresh EC2 host
⏱ 30-90 min
~$2.22-5.14/hr. Recommended: m8id.8xlarge (32 vCPU / 128 GiB / 1.9 TB NVMe).
GCP
Coming soon
⏱ —
Not yet available. Use bare metal or AWS for now.
Quick start (bare metal):
curl -fsSL https://raw.githubusercontent.com/avaleror/rodeo-cli/main/install.sh | bash
git clone https://github.com/avaleror/suse-virt-workshop.git
cd suse-virt-workshop/baremetal
rodeo doctor
rodeo up
rodeo up takes 90-150 minutes, survives SSH drops via tmux, and prints
Harvester (:8443) and Rancher (:30002) URLs when done. Full instructor steps,
and the AWS equivalent: Host setup: bare metal ·
Host setup: AWS.
Lab topology¶
KVM host (bare metal, yours) │ ├── harvester1 192.168.122.11 8 vCPU / 16 GiB / 320 GB bootstrap ├── harvester2 192.168.122.12 8 vCPU / 16 GiB / 320 GB join ├── harvester3 192.168.122.13 8 vCPU / 16 GiB / 320 GB join └── rancher 192.168.122.9 4 vCPU / 8 GiB / 60 GB Rancher Prime + K3s VIP (kube-vip): 192.168.122.10 DNAT: host:8443 → VIP:443 · host:30002 → rancher:30002
Exercises¶
Exercise 01
The Arrival
⏱ 30 min
Import Harvester into Rancher, tour both UIs, confirm storage and terminal access.
Exercise 02
The Subterranean Divide
⏱ 30 min
Map node topology, create prod/dev workspaces, tune Longhorn tiers, build the service network.
Exercise 03
The Flash Crash
⏱ 30 min
Provision algo-trader-01 with volumes, network, cloud-init, and a fixed IP under pressure.
Exercise 04
The Rising Tide
⏱ 25 min
Zero-downtime live migration of the payment gateway, then evacuate a damaged node.
Exercise 05
The Invisible Intruder
⏱ 30 min
Build a closed-loop cluster network and an overlay vault so lateral movement dies.
Exercise 06
The Unthinkable Error
⏱ 30 min
Snapshot, clone to staging, restore production, wire an NFS backup target and schedule.
Exercise 07
The Stampede
⏱ 25 min
Forge a golden VM template and stamp out a calculation fleet on demand.
Exercise 08
A New Horizon
⏱ 10 min
Recap what you mastered and where the skills go next.
Bonus (self-hosted only, no rodeo counterpart)¶
Legacy → SUSE mapping¶
| Old world (ISAware) | SUSE Virtualization |
|---|---|
| Proprietary hypervisor | KubeVirt + KVM/QEMU |
| Storage array | Longhorn (or any CSI driver) |
| Closed-source SDN | Kube-OVN + Multus |
| Command console | Rancher Prime + Harvester UI |
Component versions¶
| Component | Version |
|---|---|
| Harvester | 1.8.1 |
| Rancher Prime | 2.14.1 |
| K3s | v1.35.3+k3s1 |
| cert-manager | v1.20.1 |
| rodeo-cli | v0.14.x |