No Network Connection

Quick Diagnosis


# Check network interface status
ip link show
 
# Check IP addresses
ip addr show
 
# Test network connectivity
ping -c 4 8.8.8.8
 
# Test DNS resolution
ping -c 4 www.google.com
 
# Check NetworkManager status
nmcli general status

Wired Network

Network card not recognized


# Check Ethernet devices
lspci | grep -i ethernet
ip link show | grep -E "^[0-9]"
 
# If the network card is not shown, check the driver
sudo dmesg | grep -i eth
sudo dmesg | grep -i r8169  # Common Realtek driver

No IP address obtained


# Have NetworkManager re-request an address (dhclient is no longer shipped)
sudo nmcli device reapply enp0s3
# On systemd-networkd systems use instead:
sudo networkctl renew enp0s3
 
# Or reconnect the interface
sudo nmcli device connect enp0s3
 
# Manually configure a static IP
sudo nmcli con mod "Wired connection 1" \
  ipv4.addresses 192.168.1.100/24 \
  ipv4.gateway 192.168.1.1 \
  ipv4.dns "8.8.8.8,1.1.1.1" \
  ipv4.method manual
sudo nmcli con up "Wired connection 1"

Wireless Network

Wi-Fi adapter not detected


# Check wireless devices (iwconfig is deprecated; use iw)
iw dev
nmcli device wifi list
 
# Check if disabled by RF switch
rfkill list all
sudo rfkill unblock wifi

Cannot connect to Wi-Fi


# Scan for available networks
nmcli device wifi rescan
nmcli device wifi list
 
# Connect to Wi-Fi
nmcli device wifi connect "SSID_NAME" password "PASSWORD"
 
# View saved connections
nmcli connection show
 
# Delete a problematic connection and re-add
nmcli connection delete "CONNECTION_NAME"

Frequent disconnections


# Check if power management turned off Wi-Fi
iw dev wlan0 get power_save
 
# Disable Wi-Fi power management
sudo iw dev wlan0 set power_save off
 
# Make it permanent
echo 'options iwlwifi power_save=0' | sudo tee /etc/modprobe.d/iwlwifi-powersave.conf

DNS Issues

Symptom: IPs work, names do not


# 1. Does the IP layer work at all (bypasses DNS entirely)?
ping -c 2 223.5.5.5
 
# 2. Does name resolution fail?
getent hosts archive.ubuntu.com

If step 1 succeeds and step 2 hangs or reports that the name cannot be resolved, the problem is DNS, not the link. This is easy to misread as “no internet”, and the apt errors it produces look like the repository is unreachable.

Three commands to locate it


# 1. Which DNS server is this machine using?
resolvectl status
 
# 2. Ask that server directly (use the address from step 1)
dig +short @223.5.5.5 archive.ubuntu.com
 
# 3. Compare against a public resolver
dig +short @119.29.29.29 archive.ubuntu.com

dig comes from the dnsutils package (sudo apt install dnsutils); getent hosts <name> goes through the system resolver if you prefer not to install anything.

The three outcomes map to three causes:

What you see	Cause	Fix
The configured server does not answer (step 2 empty or timing out) while a public one works	DHCP handed out an unusable DNS, usually because the router’s own forwarder died	Change DNS, next section
Neither answers	Outbound blocked, or the upstream is down	See Firewall Issues and verify the gateway is reachable
Both answer but resolution still fails	Local cache or proxy settings	`sudo resolvectl flush-caches` and retry

Changing DNS: temporary and permanent

Temporary (gone after reboot, the usual first aid):


# Replace ens18 with your interface name; `ip -brief addr` shows it
sudo resolvectl dns ens18 223.5.5.5
sudo resolvectl flush-caches

For a permanent change, first find out what manages the interface:


nmcli -t -f NAME,DEVICE con show --active

NetworkManager (the desktop default):


sudo nmcli con mod "netplan-ens18" ipv4.dns "223.5.5.5 119.29.29.29"
sudo nmcli con mod "netplan-ens18" ipv4.ignore-auto-dns yes
sudo nmcli con up "netplan-ens18"

The connection name must match the output above. A wrong name fails with “unknown connection”, which is easy to miss and leaves everything unchanged.

netplan (edits the config file, survives reboots, recommended):


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


network:
  ethernets:
    ens18:
      dhcp4: true
      dhcp4-overrides:
        use-dns: false        # key: stop accepting DNS from DHCP
      nameservers:
        addresses: [223.5.5.5, 119.29.29.29]
  version: 2

Then apply:


sudo netplan apply

use-dns: false is the important part. Without it the broken address from DHCP is merged back in, and resolvectl status shows your server and the bad one side by side.

The host resolves, the virtual machine does not

When the host — or any other machine on the LAN — resolves fine and only one VM fails, the router is usually advertising itself as DNS while its forwarder is dead.

Confirm it:


# Run this on both the VM and the host, with your gateway address
dig +short @192.168.1.1 archive.ubuntu.com

No answer on either machine means the gateway’s forwarding is broken. Three ways out:

Fix the router’s DHCP/WAN DNS settings (recommended — repairs every device at once)
Run a resolver on the LAN (for example dnsmasq on an always-on host) and point the VMs at it
Change only this VM’s netplan to a public resolver (most local, but you repeat it per machine)

Recommended resolvers

Provider	Primary	Secondary
Alibaba Cloud	223.5.5.5	223.6.6.6
Tencent DNSPod	119.29.29.29	—
114 DNS	114.114.114.114	114.114.115.115
Google	8.8.8.8	8.8.4.4

Firewall Issues


# Check UFW status
sudo ufw status verbose
 
# Temporarily disable firewall for testing
sudo ufw disable
 
# Allow specific ports
sudo ufw allow 80/tcp
sudo ufw allow 443/tcp

Proxy Settings

System Proxy


# Set environment variables
export http_proxy="http://127.0.0.1:7890"
export https_proxy="http://127.0.0.1:7890"
export no_proxy="localhost,127.0.0.1"
 
# Write to ~/.bashrc for permanence

APT Proxy


sudo nano /etc/apt/apt.conf.d/proxy.conf


Acquire::http::Proxy "http://127.0.0.1:7890";
Acquire::https::Proxy "http://127.0.0.1:7890";

Reset Network Configuration


# Restart NetworkManager
sudo systemctl restart NetworkManager
 
# Completely reset network configuration
sudo rm /etc/NetworkManager/system-connections/*
sudo systemctl restart NetworkManager