Handbook (Operator Cheat-Sheet)

A one-screen operational reference for working inside the testbed. Canonical detail lives in the linked owners below; this page does not restate IPs, VNIs, or phase documentation (see CLAUDE.md for the ownership charter).

kubectl: K3s ships no standalone kubectl. Run from master (vagrant ssh master) with sudo k3s kubectl.

Where to find what

For	See
Interface matrix (subnets, static IPs, VXLAN VNIs)	architecture/5g-interfaces.md
Node/VM topology and IPs	architecture/overview.md
Network design (OVS, VXLAN, Multus)	architecture/network-topology.md
Deployment phases	deployment/phases.md
Dashboard access and modules	dashboard/overview.md
IAM, roles, OIDC clients	security/iam.md
External access and tunnels	security/external-access.md, deployment/external-tunnel.md
Troubleshooting	troubleshooting.md + runbooks/
Feature maturity	status.md
Roadmap	roadmap.md
Known issues	known-issues/

Quick commands

Cluster status

sudo k3s kubectl get nodes -o wide
sudo k3s kubectl get pods -A
sudo k3s kubectl -n 5g get deploy,svc,pods
sudo k3s kubectl top pods -n 5g
sudo k3s kubectl top nodes

Network

# Interfaces on a pod
sudo k3s kubectl -n 5g exec deploy/amf -- ip addr show

# Overlay
sudo ovs-vsctl show
ip -d link show | grep vxlan
sudo k3s kubectl get net-attach-def -A

Logs

# K3s
journalctl -u k3s -f
journalctl -u k3s-agent -f

# KubeEdge
sudo k3s kubectl -n kubeedge logs -l app=cloudcore -f
ssh edge "journalctl -u edgecore -f"

# 5G Core
sudo k3s kubectl -n 5g logs deploy/amf -f
sudo k3s kubectl -n 5g logs deploy/smf -f

# OVS DaemonSet
sudo k3s kubectl -n kube-system logs -l app=ds-net-setup-worker -f

Configuration files

• Ansible variables: ansible/group_vars/all.yml
• Vagrant: Vagrantfile
• 5G Core configs: ansible/phases/05-5g-core/configs/
• OVS scripts: ansible/phases/04-overlay-network/scripts/

Operations

Pod migration (UPF/MEC between cloud and edge)

upf-edge is disabled by default (replicas: 0); it is shown here only as the migration example.

sudo k3s kubectl -n 5g get pods -o wide

# Pin a deployment to a node
sudo k3s kubectl -n 5g patch deployment upf-edge \
  -p '{"spec":{"template":{"spec":{"nodeSelector":{"kubernetes.io/hostname":"worker"}}}}}'

sudo k3s kubectl -n 5g rollout status deploy/upf-edge --timeout=120s

Restart and rollout

sudo k3s kubectl -n 5g rollout restart deployment/amf
sudo k3s kubectl -n 5g rollout status deployment/amf

Backup and recovery

# Back up Ansible config and live manifests
tar -czf ansible-config-backup.tar.gz ansible/
sudo k3s kubectl get all -A -o yaml > k8s-manifests-backup.yaml
sudo ovs-vsctl show > ovs-config-backup.txt

# Restore
tar -xzf ansible-config-backup.tar.gz
sudo k3s kubectl apply -f k8s-manifests-backup.yaml

---

For focused subsystem diagnostics (NGAP, PFCP, GTP-U, OVS, Multus) see the runbooks. For deeper procedures see the topic owners in the table above.