Appendix B: Final Exam
Goal: Four tasks without instructions that tie together everything from the book. · Duration: approx. 60 min · Prerequisite: Day 0–15
Rules of Play
- You work in
~/one-lab/pruefung. The OneFlow servicenotes-servicefrom Day 9 is running. - Every task has check commands. Only when they show what is written there is the task solved.
- The solutions are in
~/one-lab/code/loesung/pruefung/.
mkdir -p ~/one-lab/pruefung
cd ~/one-lab/pruefungExercise 1: Team rot
Team rot gets a group team-rot with a user
rot-dev (password lab-rot-pass-123) and may
have exactly one VM at a time. It should be able to use the
alpine template.
Check:
B="--user rot-dev --password lab-rot-pass-123"
onetemplate instantiate alpine --name rot1 --user-inputs GRUSS=r $B
onetemplate instantiate alpine --name rot2 --user-inputs GRUSS=r $BExpected (the IDs will differ):
VM ID: 52
[one.template.instantiate] User [3] : group [101] limit of 1 reached for VMS quota in VM.
Solution
onegroup create team-rot
oneuser create rot-dev lab-rot-pass-123
oneuser chgrp rot-dev team-rot
printf 'VM = [ VMS = "1" ]\n' > quota.txt
onegroup quota team-rot quota.txtYou already shared the template, image, and security group with
everyone on Day 8 (604). That is why the first VM works
without further sharing. Clean up afterwards:
onevm terminate --hard rot1 $BSave your progress? Sign up for free
Exercise 2: Quitting time for the API
The service’s first API VM should be powered off on weekdays at 10 p.m. (UTC).
Check:
V=$(onevm list --no-header | grep api_0 | awk '{print $1}')
onevm show $V | sed -n '/^SCHEDULED ACTIONS/,/^$/p'Expected (date and remaining time will differ):
SCHEDULED ACTIONS
ID ACTION ARGS SCHEDULED REPEAT END STATUS
1 poweroff - 11/13 22:00 Weekly 1,2,3,4,5 None Next in 1.04 days
Solution
onevm poweroff $V --schedule "11/13 22:00" --weekly 1,2,3,4,5Expected:
VM 30: poweroff scheduled at 2026-11-13 22:00:00 +0000
Every onevm action can be scheduled this way, like the
backup on Day 7. 1,2,3,4,5 is Monday through Friday.
Save your progress? Sign up for free
Exercise 3: How many are running?
Write a Python script zustand.py that uses the XML-RPC
API to count how many VMs are in which state, with names instead of
numbers.
Check:
python3 zustand.pyExpected (the number depends on what is currently running):
ACTIVE 3
Solution
cat > zustand.py <<'EOF'
#!/usr/bin/env python3
"""Zählt die VMs pro Zustand über die XML-RPC-API."""
import collections
import xml.etree.ElementTree as ET
import xmlrpc.client
SESSION = "oneadmin:lab-one-pass-123"
NAMEN = {"1": "PENDING", "3": "ACTIVE", "8": "POWEROFF", "9": "UNDEPLOYED"}
one = xmlrpc.client.ServerProxy("http://localhost:2633/RPC2")
ok, body, code = one.one.vmpool.info(SESSION, -2, -1, -1, -1)[:3]
if not ok:
raise SystemExit(f"Fehler {code}: {body}")
zahl = collections.Counter(vm.findtext("STATE") for vm in ET.fromstring(body).findall("VM"))
for state, n in sorted(zahl.items()):
print(NAMEN.get(state, state), n)
EOFFor states, the plain pool data is enough; you do not need
infoextended from Day 13 here.
Save your progress? Sign up for free
Exercise 4: Room for logs
The first API VM gets an additional, non-persistent disk
logs with 256 MB while it is running.
Check:
onevm show $V | sed -n '/^VM DISKS/,/^$/p' | grep logs
IP=$(onevm show $V | grep -o 'ETH0_IP="[0-9.]*' | cut -d'"' -f2)
ssh $IP "grep sd /proc/partitions"Expected:
2 default sdb logs -/256M file NO
8 0 262144 sda
Solution
oneimage create --name logs --type DATABLOCK --size 256 --datastore default
onevm disk-attach $V --image logsAs on Day 7, the disk is called sda inside the VM and
sdb for OpenNebula.
Save your progress? Sign up for free
You made it
When all four exercises pass their check commands, you have completed all the drills. You can:
- install OpenNebula, name its parts, and read their logs,
- create images, datastores, templates, and networks and understand when data survives,
- configure VMs via the context and build appliances and your own images from them,
- master the lifecycle, snapshots, and backups,
- separate teams with permissions and quotas,
- build and scale services with OneFlow,
- use OneGate, hooks, OpenTofu, and the APIs for automation,
- operate a host, back it up, and plan an upgrade.
Clean up when you no longer need the lab: log out and delete the lab VM in your hypervisor. Everything OpenNebula created lives inside it.