Configure a static IP on elementary OS for network stability

A stable local address makes an elementary OS workstation easier to reach from other devices and easier to host services on. Learning how to configure a static IP address on elementary OS for network stability is particularly useful for a home-office computer, network-attached storage, a shared printer or a media server. The aim is an address that stays predictable through reboots, reconnects and sleep, while still fitting the router’s existing network.

Current elementary OS releases use NetworkManager to manage wired connections, and the nmcli command-line client provides precise control. A fixed IPv4 address can be entered in the graphical controls, saved inside a connection profile or implemented as a DHCP reservation in the router. A reservation is often the tidier household option because the router keeps assigning the same address, while direct configuration gives the desktop control over its own local identity.

This guide covers a fixed private IPv4 address on the LAN, not a guaranteed public address. The distinction matters in Australia, where many NBN subscribers receive a dynamic public IPv4 and access the internet through carrier-grade NAT. A static address on the elementary OS host can prevent DHCP conflicts and keep local services consistent; it does not stop an ISP changing its public address or remove NAT limitations.

Plan the address before changing the desktop

Begin with the router’s current network plan rather than an arbitrary number. DHCP assigns a temporary lease, a router reservation ties a device’s MAC address to a fixed lease, and a directly configured IPv4 address is stored in NetworkManager. All three can provide a predictable endpoint, but they differ in administration and recovery.

Choose a subnet that matches the router, then reserve an unused host address inside it. For a router at 192.168.1.1 with a /24 subnet, a possible workstation address is 192.168.1.50, the subnet mask is 255.255.255.0, and the default gateway is 192.168.1.1. DNS can come from the router or use public resolvers such as 1.1.1.1 and 9.9.9.9. Record the address, prefix, gateway, DNS servers and connection profile before editing anything. These are examples, not defaults that should be copied blindly. A mismatched gateway or subnet can leave the computer unable to reach other devices on the LAN.

Check that the chosen address sits outside the router’s DHCP range, or exclude it from that range. Otherwise, the desktop and a leased device can eventually receive the same address. Common Australian home setups in Sydney, Melbourne and Brisbane combine an NBN-supplied router with personal Wi-Fi equipment for streaming, printing and home-office use. Make one router the only DHCP server on the network to prevent overlapping leases, competing gateways and unexpected DNS changes.

Use the graphical network controls

Open Settings, select Network, and choose the wired connection used by the desktop. Current elementary OS releases expose NetworkManager connection options, although labels vary between versions. Look for an IPv4 tab under Edit or Configure, and note the active profile name if one is shown. Keep access to the local screen and keyboard in case a change interrupts the network.

Change the IPv4 method from Automatic to Manual. Enter the address and subnet prefix or mask, add the default gateway, and specify DNS servers if you want explicit resolvers rather than the router’s DHCP-supplied values. Leave IPv6 on Automatic unless you have a separate IPv6 plan; an IPv4 setting does not require disabling IPv6. Apply the changes, reactivate the wired profile, and confirm that the router lists the device with the expected address. An error message about an existing address usually means the selected IP is already allocated somewhere on the network.

Graphical editing is convenient, but some releases hide advanced fields or expose only part of the connection profile. If the manual IPv4 method is unavailable, or the gateway and DNS fields are missing, use the terminal instead. Do not create a second profile with the same name merely to experiment, as NetworkManager may activate the wrong connection after a reboot. Updating the existing active profile also avoids leaving conflicting configurations behind.

Configure the connection with nmcli

Start by identifying both the active connection profile and the Ethernet device. The connection name shown by nmcli is what later commands expect, while the device name can change if hardware or USB adapters are rearranged. Work at the desktop rather than through a remote SSH session until the new address has been confirmed.

nmcli connection show
nmcli device status

Replace the example profile name and local addresses with your own, then modify the existing connection. A slash after the address expresses the prefix, so /24 is equivalent to the 255.255.255.0 mask. The comma-separated DNS list is stored on the profile and takes precedence over automatically discovered resolvers when automatic DNS is ignored.

nmcli connection modify "Wired connection 1" \
  ipv4.method manual \
  ipv4.addresses "192.168.1.50/24" \
  ipv4.gateway "192.168.1.1" \
  ipv4.dns "192.168.1.1,1.1.1.1" \
  ipv4.ignore-auto-dns no

Activating the profile applies the new address, route and resolver settings. A brief disconnect is normal while the wired link renegotiates. If the address is outside the DHCP pool, the desktop can retain the same local identity across router renewals and power cycles. This changes only the IPv4 configuration on the elementary OS machine; the ISP’s public address remains under the provider’s control.

nmcli connection up "Wired connection 1"

Verify the connection and recover from errors

Check the resulting address, default route and name resolution separately. Replace the sample gateway with your router’s address. The first command below should show the manual address on the wired interface, and the route list should contain the chosen gateway as its default.

ip -brief address
ip route
ping -c 3 192.168.1.1
ping -c 3 1.1.1.1
ping -c 3 one.one.one.one

A failed router ping points towards the local link, subnet or address. A successful router ping followed by a failed external ping suggests an ISP, WAN or firewall problem. A numeric address that works while a hostname fails points towards DNS. Review recent NetworkManager events with the journal command below, and check any firewall rules rather than disabling protection wholesale. Device-specific details can also be inspected with nmcli device show followed by the interface name. The resolver status is useful when a connection appears active but names cannot be resolved.

sudo journalctl -u NetworkManager -b --no-pager

If the change made the machine unreachable, use the router’s administration page or return to a local terminal and restore DHCP. These commands clear the manual values from the named profile and request a fresh lease. Keep the router’s address range unchanged during recovery, and keep a USB keyboard and display available on remotely managed machines.

nmcli connection modify "Wired connection 1" \
  ipv4.method auto \
  ipv4.addresses "" \
  ipv4.gateway "" \
  ipv4.dns "" \
  ipv4.ignore-auto-dns no
nmcli connection up "Wired connection 1"

Keep the local address stable over time

For most Australian homes, a DHCP reservation in the NBN or personal router is the easiest long-term arrangement. The desktop keeps using automatic addressing while the router consistently maps the computer’s MAC address to the selected IP. A direct static configuration remains useful for an appliance-style host, a router without reservation features or a deliberate network design with separate address pools.

Document the chosen address and avoid assigning it to another device. Reboot the router, reconnect the desktop and test after waking it from sleep, since these events commonly expose lease conflicts. If an ISP supplies the modem, confirm its DHCP service is disabled when a personal router handles the network. Australian Consumer Law within the Competition and Consumer Act 2010 still governs the accuracy of advertised router functions, but it does not require every ISP to provide a static public IP.

Public inbound services need separate planning. The Australian market includes NBN plans from Telstra, Optus, TPG, Aussie Broadband and other providers, with static public IPv4 generally treated as an optional feature rather than a default. Many consumer connections remain behind carrier-grade NAT, while IPv6 availability and port-forwarding rules vary by provider and plan. Check the current agreement rather than assuming a local fixed address solves remote access. For broader technology context beyond elementary OS, IT News Corner provides related reporting.

The practical takeaway is to keep one authoritative DHCP server, reserve or configure one non-conflicting local address, and retain a tested DHCP fallback. These controls keep the elementary OS host predictable across daily use and make future ISP changes easier to handle.