How to fix fatal: Cannot bind any address. in SSH

First, verify that sshd isn't already listening on the port:

sudo netstat -antp | grep :22
# or with ss (modern systems):
sudo ss -antp | grep :22

Look for LISTEN state on port 22 (or your configured SSH port). If sshd is already running, restart it cleanly:

sudo systemctl stop ssh
sudo systemctl start ssh
sudo systemctl status ssh

If stopping hangs, force it:

sudo pkill -9 sshd
sudo systemctl start ssh

If another process is listening on port 22:

sudo lsof -i :22

Or find what's listening on the SSH port:

sudo ss -tlnp | grep :22

If it's an old sshd process that didn't shut down cleanly, kill it:

sudo pkill sshd
# Wait a moment, then restart
sudo systemctl start ssh

If it's a different service, either:
- Change the SSH port in /etc/ssh/sshd_config (set Port 2222, for example)
- Or stop/disable the conflicting service

Open /etc/ssh/sshd_config and check the AddressFamily setting:

sudo nano /etc/ssh/sshd_config

Look for the AddressFamily line. If it's set to inet6 only, change it:

# Old (broken):
AddressFamily inet6

# New (fixed):
AddressFamily any

Or comment it out to use the default (which accepts both IPv4 and IPv6):

# AddressFamily any

Then check the ListenAddress lines. A common mistake is:

# This fails if IPv6 is not fully enabled:
ListenAddress ::

# Better: listen on both
ListenAddress 0.0.0.0
ListenAddress ::

Save and validate the config:

sudo sshd -t

If this shows no errors, restart sshd:

sudo systemctl restart ssh

If sshd fails only at boot but works after manually restarting, the network isn't ready when sshd starts.

For systemd systems, create an override to make sshd wait for the network:

sudo systemctl edit ssh

Add these lines in the [Unit] section:

[Unit]
After=network-online.target
Wants=network-online.target

Save (Ctrl+X in nano). The file is auto-saved and the service is reloaded.

For init.d systems (older Linux), edit /etc/init.d/ssh or the init script to start sshd after network:

# Look for the start-stop-daemon call and ensure it runs after network initialization
# For OpenRC or similar, you may need to adjust the runlevel or add a dependency

Reboot to verify:

sudo reboot
# After reboot, verify SSH is running
sudo systemctl status ssh

If you changed the SSH port (non-standard port), SELinux may block binding. Check if SELinux is enabled:

getenforce

If it shows "Enforcing" or "Permissive", you need to allow sshd on your port:

# For a custom port (e.g., 2222):
sudo semanage port -l | grep ssh
# Should show: ssh_port_t = 22

# Add your port:
sudo semanage port -a -t ssh_port_t -p tcp 2222

# Verify:
sudo semanage port -l | grep ssh

Then restart sshd:

sudo systemctl restart ssh

If you don't have semanage, install it:

# On RHEL/CentOS/Fedora:
sudo yum install policycoreutils-python-utils

# On Ubuntu (rarely needed):
sudo apt install selinux-policy-dev

If sshd_config has a ListenAddress for an IP that doesn't exist on the system:

sudo nano /etc/ssh/sshd_config

Check all ListenAddress lines:

# Problem: This IP doesn't exist on the system
ListenAddress 192.168.100.50

# Fix: Use a real IP or listen on all
ListenAddress 0.0.0.0         # All IPv4
ListenAddress ::              # All IPv6

To list all IPs on your system:

ip addr show
# or
ifconfig

Update sshd_config to use only IPs that exist, or use 0.0.0.0 and :: to listen on all interfaces.

Test the config:

sudo sshd -t

Restart:

sudo systemctl restart ssh

After applying any fix, verify sshd is running and listening:

# Check service status
sudo systemctl status ssh

# Verify it's listening
sudo ss -tlnp | grep ssh
# Should show: tcp LISTEN 0 128 0.0.0.0:22 0.0.0.0:* users:(("sshd",pid=1234,fd=3))

# Test SSH connection from another machine
ssh -v user@your-server-ip

If SSH still won't start, check the sshd logs for more details:

sudo journalctl -u ssh -n 20
# or
sudo tail -f /var/log/auth.log | grep sshd

Look for specific error messages beyond "Cannot bind any address" to identify the exact cause.