Cloning Ubuntu 24.04 VMs on Proxmox Without Conflicts

Overview¶

This guide documents the process of cloning a fully configured Ubuntu 24.04 "golden image" VM on the same Proxmox hypervisor while ensuring the clone operates independently without IP address conflicts or other network issues.

Important: This guide covers cloning a regular, fully-configured VM (not a Proxmox template). The base VM remains a working VM that you can continue to update and maintain, then clone again in the future when you need new VMs with your standard configuration.

Prerequisites¶

• A working Ubuntu 24.04 VM on Proxmox configured with your standard tools (SSH, Docker, Atuin, etc.)
• This base/golden image VM should be shut down temporarily for cloning
• Access to Proxmox web interface or command line
• Understanding of your network IP addressing scheme

Step 1: Prepare the Golden Image VM for Cloning¶

Before cloning, shut down your base VM to ensure a consistent copy.

1.1 Shut Down the Base VM¶

From Proxmox host CLI (replace 100 with your VM ID):

qm shutdown 100

Or from within the VM:

sudo shutdown -h now

Verify: In Proxmox web interface, confirm VM status shows "stopped"

1.2 Optional: Update the Golden Image¶

Before cloning, you may want to ensure your golden image has the latest updates.

SSH into the base VM before shutting it down:

sudo apt update

sudo apt upgrade -y

sudo apt autoremove -y

Then shut down:

sudo shutdown -h now

Note: You can skip cleaning operations like clearing machine-id or bash history since this is a working VM you'll continue to use and update.

Step 2: Clone the VM in Proxmox¶

2.1 Using Proxmox Web Interface¶

1. Navigate to your Proxmox datacenter view
2. Select the base/golden image VM (e.g., VM 100)
3. Right-click and select "Clone" or click the "More" button and select "Clone"
4. Configure clone settings:
5. VM ID: Choose a new unique ID (e.g., 101)
6. Name: Give it a descriptive name (e.g., ubuntu-docker-host)
7. Mode: Select "Full Clone" (not linked clone)
8. Target Storage: Select appropriate storage (usually same as source)
9. Click "Clone" to begin the process

Verify: Wait for cloning to complete (this may take several minutes), then confirm the new VM appears in your VM list.

2.2 Using Proxmox Command Line¶

Clone VM 100 to new VM 101 with full clone:

qm clone 100 101 --name ubuntu-docker-host --full

Check clone status:

qm status 101

Verify: Run qm list to see both VMs in the list (both should show "stopped").

Step 3: Start VMs in Correct Order¶

CRITICAL: After cloning, you must start the VMs in the correct order to prevent IP address conflicts.

NOTE: Any VMs that are shutdown should be started before bringing the new VM online or else these VMs might lose their existing IP address if they are not set with static addresses and use your DHCP server instead. Just saves time later tidying things up if IP addresses do "jump".

3.1 Start the Original/Golden Image VM First¶

This is essential to ensure your original VM reclaims its IP address (especially important with DHCP).

From Proxmox host:

qm start 100

Or use web interface: select VM 100, click "Start"

Verify:

Wait 30-60 seconds for boot, then check it's running:

qm status 100

Should show: status: running

Optional verification: SSH into the original VM to confirm it has its expected IP:

ssh user@192.168.1.100

ip addr show

3.2 Now Start the Cloned VM¶

Only after the original VM is fully booted and has claimed its IP address, start the clone.

From Proxmox host:

qm start 101

Or use web interface: select VM 101, click "Start"

Verify:

qm status 101

Should show: status: running

Step 4: Configure the Cloned VM¶

Now that both VMs are running, you need to access the clone via console to discover its IP address and reconfigure it.

4.1 Access Clone via Proxmox Console¶

1. In Proxmox web interface, select the cloned VM (101)
2. Click "Console" to open no VNC console
3. Log in with the same credentials as your original VM

Note: You cannot SSH yet because you don't know the clone's IP address.

4.2 Discover the Clone's IP Address¶

Check the IP address assigned to the clone:

ip addr show

Look for the inet line under your network interface (usually ens18). Example output:

inet 192.168.1.152/24 brd 192.168.1.255 scope global dynamic ens18

Note this IP address - you'll use it for SSH in the next steps.

Verify: The clone should have a different IP from your original VM (192.168.1.100 in our example).

4.3 Change Hostname¶

Check current hostname (will be same as original VM):

hostnamectl

Set new hostname:

sudo hostnamectl set-hostname ubuntu-docker-host

Edit /etc/hosts to match:

sudo nano /etc/hosts

Update /etc/hosts:

127.0.0.1       localhost
127.0.1.1       ubuntu-docker-host

# The following lines are desirable for IPv6 capable hosts
::1     localhost ip6-localhost ip6-loopback
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters

Verify:

hostnamectl

Should show new hostname.

hostname

Should also show new hostname.

4.4 Optional: Configure Static IP¶

If you want to assign a specific static IP to the clone (instead of using DHCP), update netplan.

Check current network configuration:

cat /etc/netplan/00-installer-config.yaml

Edit netplan configuration:

sudo nano /etc/netplan/00-installer-config.yaml

Example netplan for static IP:

network:
  version: 2
  ethernets:
    ens18:
      addresses:
        - 192.168.1.105/24  # Change to desired static IP
      routes:
        - to: default
          via: 192.168.1.1
      nameservers:
        addresses:
          - 192.168.1.1
          - 8.8.8.8

Apply netplan changes:

sudo netplan apply

Verify new IP:

ip addr show

Verify: Test connectivity:

ping -c 4 192.168.1.1

ping -c 4 8.8.8.8

nslookup google.com

Note: If you assign a static IP, you may lose console connection briefly. Reconnect via console if needed.

4.5 Regenerate Machine-ID (Important!)¶

The cloned VM will have the same machine-id as the original, which can cause issues with systemd and other services.

Check current machine-id:

cat /etc/machine-id

Remove old machine-id files:

sudo rm /etc/machine-id

sudo rm /var/lib/dbus/machine-id

Regenerate new unique machine-id:

sudo systemd-machine-id-setup

Create dbus symlink:

sudo ln -s /etc/machine-id /var/lib/dbus/machine-id

Verify new machine-id was created:

cat /etc/machine-id

Verify: The machine-id should be a new 32-character hex string, different from the original VM.

Note: Some services may need to be restarted after changing machine-id. A reboot is recommended after all configuration changes.

4.6 Clear Cached SSH Fingerprint on Management Laptop¶

When you SSH to the cloned VM from your management laptop, (sometimes) you'll get a warning because the IP address now has different SSH host keys than before.

From your management laptop (lpt-hp), remove the old cached fingerprint (replace with the clone's IP):

ssh-keygen -R 192.168.1.152   #the new VM IP address

Verify: Now SSH into the clone normally:

ssh user@192.168.1.152   #the new VM IP address

You'll be prompted to accept the new fingerprint. Type yes to accept.

4.7 Reboot to Apply All Changes¶

After making all configuration changes, reboot the cloned VM:

sudo reboot

Verify: After reboot, SSH into the clone using its IP address.

From your management laptop:

ssh user@192.168.1.152

Verify hostname:

hostname

Verify machine-id is unique:

cat /etc/machine-id

Step 5: Final Verification Steps¶

5.1 SSH Access from Management Laptop¶

SSH to the cloned VM using its IP:

ssh user@192.168.1.152

Verify hostname is correct:

hostname

Should show: ubuntu-docker-host

Verify machine-id is unique:

cat /etc/machine-id

Should be different from original VM.

Check SSH fingerprint:

ssh-keygen -lf /etc/ssh/ssh_host_ed25519_key.pub

Should be different from original VM.

5.2 Network Connectivity Tests¶

Test DNS resolution:

nslookup google.com

Test outbound connectivity:

curl -I https://www.google.com

Check listening services:

sudo ss -tlnp | grep :22

5.3 Verify No IP Conflicts¶

From your management laptop, verify both VMs are accessible.

Ping original/golden image VM:

ping -c 4 192.168.1.100

Ping cloned VM:

ping -c 4 192.168.1.152

SSH to original VM:

ssh user@192.168.1.100

hostname

Should show original hostname.

exit

SSH to cloned VM:

ssh user@192.168.1.152

hostname

Should show new hostname.

exit

Verify: Both VMs should respond without conflicts or timeouts.

5.4 Verify Installed Services¶

Check Docker (from your golden image):

sudo systemctl status docker

sudo docker ps

Check SSH:

sudo systemctl status ssh

Check Atuin (if installed):

systemctl --user status atuin

List all enabled services:

systemctl list-unit-files --state=enabled

5.5 Update LibreNMS or Monitoring (if applicable)¶

If you use LibreNMS or other monitoring:

1. Add the new cloned VM as a device with its new hostname and IP
2. Verify SNMP or other monitoring agents are working
3. Update any documentation or inventory systems

Step 6: Golden Image Maintenance¶

6.1 Keeping Your Golden Image Updated¶

Your golden image VM (ID 100) should be periodically updated.

SSH into the golden image VM:

ssh user@192.168.1.100

Update packages:

sudo apt update

sudo apt upgrade -y

sudo apt autoremove -y

Update Docker images if you have base images:

sudo docker image prune -a

Restart if kernel was updated:

sudo reboot

Recommendation: Update your golden image monthly or before creating new clones.

6.2 Document Your VM Inventory¶

Keep track of your VMs and their purposes:

6.3 When to Create a New Golden Image¶

Consider creating a fresh golden image when: - Ubuntu releases a new LTS version - You've made so many changes to the current golden image that it's becoming unwieldy - You want to standardize on a different set of base tools - The current golden image has accumulated technical debt

Best Practices¶

1. Always shut down the golden image VM before cloning to ensure a consistent snapshot
2. Always start the original VM first after cloning to ensure it reclaims its IP address
3. Always use full clones for production VMs (not linked clones)
4. Use Proxmox console to discover the clone's IP address before attempting SSH
5. Change hostname immediately after first boot of the clone
6. Regenerate machine-id and SSH keys to ensure each VM is unique
7. Document your IP addressing scheme to avoid confusion
8. Keep a VM inventory spreadsheet/document tracking all clones and their purposes
9. Update your golden image regularly but test updates before cloning
10. Use descriptive VM names that indicate purpose (e.g., ubuntu-docker-host, ubuntu-monitoring)
11. Consider DHCP reservations in your router for cloned VMs you want to keep long-term
12. Test SSH connectivity from your management laptop before considering the clone complete
13. Update monitoring systems (LibreNMS, etc.) when adding new VMs
14. Take snapshots of important cloned VMs after initial configuration

Troubleshooting¶

Wrong VM Started First¶

If you accidentally started the clone before the original:

1. Shut down BOTH VMs immediately:

   qm shutdown 100
   

   qm shutdown 101
   

2. Wait for both to fully stop

3. Start the original (100) first:

   qm start 100
   

4. Wait 60 seconds for it to fully boot and claim its IP

5. Then start the clone (101):

   qm start 101
   

IP Address Conflict Detected¶

If you see ARP conflicts or both VMs have the same IP:

1. Check which VM has which IP.

Access each via Proxmox console.

VM 100 console:

ip addr show

VM 101 console:

ip addr show

1. If the clone has the original's IP, restart both VMs in correct order (see above)
2. If using static IPs, immediately configure the clone with a different IP via console

Clone Won't Start¶

Check VM configuration:

qm config 101

Check Proxmox logs:

tail -f /var/log/syslog | grep kvm

Verify disk is accessible:

qm list

Can't SSH After Cloning¶

• Verify IP address with ip addr show in console
• Check SSH service: sudo systemctl status ssh
• Verify firewall: sudo ufw status
• Check SSH key fingerprint changed

Machine-ID Issues¶

If services complain about duplicate machine-id:

sudo rm /etc/machine-id

sudo rm /var/lib/dbus/machine-id

sudo systemd-machine-id-setup

sudo systemctl reboot

Advanced: Scripting the Post-Clone Configuration¶

For frequent cloning, you can create a script to automate the post-clone configuration:

#!/bin/bash
# post-clone-config.sh
# Run this script on a newly cloned VM

NEW_HOSTNAME="$1"

if [ -z "$NEW_HOSTNAME" ]; then
    echo "Usage: $0 <new-hostname>"
    exit 1
fi

echo "Configuring cloned VM with hostname: $NEW_HOSTNAME"

# Change hostname
sudo hostnamectl set-hostname "$NEW_HOSTNAME"

# Update /etc/hosts
sudo sed -i "s/127.0.1.1.*/127.0.1.1\t$NEW_HOSTNAME/g" /etc/hosts

# Regenerate machine-id
sudo rm -f /etc/machine-id /var/lib/dbus/machine-id
sudo systemd-machine-id-setup
sudo ln -sf /etc/machine-id /var/lib/dbus/machine-id

# Regenerate SSH host keys
sudo rm -f /etc/ssh/ssh_host_*
sudo ssh-keygen -A
sudo systemctl restart ssh

echo "Configuration complete. Current IP address:"
ip addr show | grep "inet " | grep -v "127.0.0.1"

echo ""
echo "Please reboot the VM to ensure all changes take effect:"
echo "  sudo reboot"

Save this script to your golden image VM at /usr/local/bin/post-clone-config.sh and make it executable:

sudo chmod +x /usr/local/bin/post-clone-config.sh

After cloning and starting the VMs in correct order, run this script via console:

sudo /usr/local/bin/post-clone-config.sh ubuntu-new-server

Related Documentation¶

• Ubuntu Server Guide: Net plan
• Proxmox VE: QEMU/KVM Virtual Machines
• Ubuntu: Network Configuration

Conclusion¶

Cloning a fully-configured Ubuntu VM on Proxmox is an efficient way to deploy new servers with your standard tools and configurations. The process is straightforward but requires careful attention to the startup sequence and post-clone configuration to avoid IP conflicts and ensure each VM has a unique identity.

Critical Steps Summary¶

1. Shut down the golden image VM before cloning
2. Clone using Proxmox (full clone, not linked)
3. Start the original VM first - this is critical for IP address management
4. Then start the clone - it will get a new IP via DHCP
5. Use console to discover the clone's IP address
6. Reconfigure the clone: hostname, machine-id, SSH keys
7. Verify both VMs are accessible and functioning independently

Key Principles¶

• The original/golden image VM always starts first after cloning
• Always use full clones for independent VMs
• Use console access when you don't know the IP yet
• Change hostname and regenerate unique identifiers immediately
• Verify connectivity before considering the clone production-ready
• Keep your golden image updated and well-documented

By following this methodical process and verifying at each stage, you can reliably deploy new VMs from your golden image while maintaining network stability and avoiding conflicts.

Last updated: 2026-01-25