Changing an SSH port requires changes to the listener and the network access rules. This example uses a service-managed RHEL-family server and an example new port of 2222.

Check the server first

Use an administrator account, keep the current session open and confirm access to a recovery console. Check that the proposed port is unused:

sudo ss -ltnp

This procedure assumes sshd.service manages SSH directly. It does not describe Ubuntu’s socket-activated setup.

Allow the new port

If firewalld is the active firewall, add the port to the zone handling your administration connection, replacing public where necessary:

sudo firewall-cmd --get-active-zones
sudo firewall-cmd --zone=public --add-port=2222/tcp
sudo firewall-cmd --permanent --zone=public --add-port=2222/tcp

The first add changes the running rules; the second preserves the allowance across a firewall reload. Update any provider firewall too. Do not enable firewalld alongside a different firewall manager just for this procedure.

On an SELinux-enforcing server, inspect existing port assignments:

sudo semanage port -l

If the new port is unassigned, add the SSH assignment:

sudo semanage port -a -t ssh_port_t -p tcp 2222

If it already belongs to another type, investigate the conflict rather than reassigning it blindly. The semanage command requires the distribution’s policy management package.

Configure the listener

  1. Back up /etc/ssh/sshd_config and inspect its included files.

  2. Edit the applicable configuration to retain the existing port temporarily and add the new one:

    Port 22
    Port 2222
  3. Check any ListenAddress directives for an explicit conflicting port.

  4. Validate and reload:

    sudo /usr/sbin/sshd -t
    sudo systemctl reload sshd
    sudo systemctl status sshd

Do not reload if validation reports errors.

Verify a new connection

From another terminal:

ssh -p 2222 [email protected]

Use the real server details. Confirm login and administrator access before removing the old port from the configuration and firewalls. Validate and test again after cleanup while keeping the recovery route available.