Bonus: The Final Showdown¶
Time: 25 min
Previous: Exercise 7: The Stampede
Next: Exercise 8: A New Horizon
This is a self-hosted-only bonus, not one of the eight chapters. suse-virt-rodeo (the Instruqt track this workshop mirrors) has exactly 8 chapters, ending at "A New Horizon." There's no "Final Showdown" chapter there. This exercise exists here because Harvester's real Migration feature (importing VMs from a legacy hypervisor) is worth trying hands-on if you have a vSphere/ISAware-compatible source to point it at, but it was never built into the graded rodeo track. Skip straight to Exercise 8: A New Horizon if you don't.
The legacy vendor wants forty percent more for the renewal. Sarah declines. The last critical workload, legacy-ledger-vm, must land on SUSE Virtualization while the old world is still running.
Tip: This lab has no legacy hypervisor to migrate from. That infrastructure doesn't exist in either this repo or suse-virt-rodeo's own automation. Below you walk the real Migration UI path, then complete the same verification steps on a stand-in ledger VM so the day-one-ops skills still stick.
B.1 Look at the migration bridge¶
In Harvester: Advanced → Migration → Sources.
The list is empty unless you point a source at a real vSphere/ISAware-compatible endpoint. Neither this repo nor suse-virt-rodeo's own automation ships one. Open Create and read the fields (source type, endpoint, credentials) so you know what production needs. Cancel without saving if you have no legacy cluster.
B.2 How extraction works (when a source exists)¶
Migrations → Create (reference, use when a source is Ready):
| Setting | Value |
|---|---|
| Source Type | ISAware (or the type your source uses) |
| Source VM | legacy-ledger-vm |
| Target Network | prod/service (or mgmt / default lab network) |
The importer copies disks, converts them to Longhorn volumes, and registers a native Harvester VM. Progress is visible on the Migrations page.
B.3 Workshop path: stand in the migrated ledger¶
Create (or reuse) a VM that represents the extracted workload:
| Field | Value |
|---|---|
| Name | legacy-ledger-vm |
| Namespace | prod |
| Network | prod/service |
| Image / SSH | same foundations as Exercise 2 |
Wait until it is Running. Open Console. The migrated ledger is alive on the new fabric.
B.4 Enable guest telemetry¶
SSH into the VM and start the QEMU guest agent:
Back in the UI, the VM should report richer guest info (IP, memory tools) once the agent is up.
B.5 Treat the refugee like a citizen¶
Prove day-one parity on the new platform:
- Take a snapshot named
post-migration-baseline. - Live-migrate
legacy-ledger-vmto another node (Exercise 4 skills). - Optional CLI:
kubectl get vm -n prod | grep legacy-ledger
kubectl get virtualmachineimports -A # empty unless a real import ran
On ISAware, live migration was often a licensed add-on. Here it is standard.