Connecting to Access Server with Linux
Connecting to OpenVPN Access Server from Linux requires a client program. It will capture the traffic you wish to send through the OpenVPN tunnel, encrypting it and passing it to the OpenVPN server. And of course, the reverse, to decrypt the return traffic.
Linux Packages Discussed
OpenVPN Access Server | openvpn-as |
OpenVPN 3 Linux Client | openvpn3 |
OpenVPN open source | openvpn |
OpenVPN 3 Linux Client
The OpenVPN 3 Linux project is a new client built on top of the OpenVPN 3 Core Library. This client is the official OpenVPN Linux Client program. You can find an overview of the features, frequently asked questions, and instructions on installing the openvpn3 package on our OpenVPN 3 for Linux site.
After following the instructions there to install the client, you’ll need a connection profile. This is a file generated by your OpenVPN Access Server installation for your specific user account. It contains the required certificates and connection settings. Go to the Client web interface of your Access Server (the main address, not the /admin portion). Log in with your user credentials. You will be shown a list of files available to download. Pick the user-locked profile or the auto-login profile, and you will be sent a client.ovpn file. Save this file to your Linux operating system.
Once you’ve moved the file to your Linux system, you can import it.
openvpn3 config-import —config $
You can start a new VPN session:
openvpn3 session-start —config $
You can manage a running VPN session:
And so on. More details can be found here: OpenVPN3Linux.
OpenVPN open source OpenVPN CLI program
The open source project client program can also connect to the Access Server. The package is available in most distributions and is known simply as openvpn. It supports the option to connect to multiple OpenVPN servers simultaneously, and it comes with a service component that can automatically and silently start any auto-login profiles it finds in the /etc/openvpn folder, even before a user has logged in. This service component can be set to automatically start at boot time with the tools available in your Linux distribution if supported. On Ubuntu and Debian, when you install the openvpn package, it is automatically configured to start at boot time.
To install the OpenVPN client on Linux, it is possible in many cases to just use the version that is in the software repository for the Linux distribution itself. If you run into any connectivity problems when using outdated software, it may be due to a possible lack of support for higher TLS versions in older versions of OpenVPN. Follow the instructions found on the open source openvpn community wiki if you wish to install the OpenVPN client on your Linux system.
After installing, you will need a connection profile. This is a file generated by your OpenVPN Access Server installation for your specific user account. It contains the required certificates and connection settings. Go to the Client web interface of your Access Server (the main address, not the /admin portion). Log in with your user credentials. You will be shown a list of files available to you for download. Pick the user-locked profile or the auto-login profile, and you will be sent a client.ovpn file. Save this file to your Linux operating system somewhere. OpenVPN Access Server supports server-locked, user-locked, and auto-login profiles, but the OpenVPN command line client is only able to connect with user-locked or auto-login connection profiles.
We are assuming you are going to start the connection through either the command line as a root user, or via the service daemon. If you want unprivileged users to be able to make a connection, take a look at the community wiki for more information on how to implement that. Here we are going to focus on the simplest implementation; run the connection as root user directly, or via the service daemon.
Start a connection with an auto-login profile manually:
openvpn --config client.ovpn
Start a connection with a user-locked profile manually:
openvpn --config client.ovpn --auth-user-pass
If you use Google Authenticator or another extra factor authentication, add the auth-retry parameter:
openvpn --config client.ovpn --auth-user-pass --auth-retry interact
To start an auto-login connection via the service daemon, place client.ovpn in /etc/openvpn/ and rename the file. It must end with .conf as file extension. Ensure the service daemon is enabled to run after a reboot, and then simply reboot the system. The auto-login type profile will be picked up automatically and the connection will start itself. You can verify this by checking the output of the ifconfig command; you should see a tun0 network adapter in the list.
One major feature that is missing with the command line client is the ability to automatically implement DNS servers that are pushed by the VPN server. It is possible, but it requires you to install a DNS management program such as resolvconf or openresolv, and it may or may not clash with existing network management software in your OS. The idea here, however, is that you use a script that runs when the connection goes up, and when it goes down, that uses resolvconf or openresolv to implement the DNS servers for you. The reason why this client is not able to manage it completely by itself is mainly because in an operating system like Windows, Macintosh, Android, or iOS, there is already an established single method of handling DNS management. It is therefore easy for us to create a software client for those operating systems that already knows how to handle DNS. But Linux is available in so many variations and also supports different programs and methods of implementing DNS servers, and so it was only reasonable to leave built-in DNS support out of the OpenVPN program and instead to provide, where possible, a script that handles DNS implementation. Such a script could even be written by yourself to do whatever tasks are necessary to implement the DNS servers in your unique situation.
Fortunately on Ubuntu and Debian, for example, there is the /etc/openvpn/update-resolv-conf script that comes with the openvpn package that handles DNS implementation for these operating systems. You need only to activate the use of these by following the instructions:
Open your client.ovpn file in a text editor:
At the very bottom simply add these lines:
script-security 2 up /etc/openvpn/update-resolv-conf down /etc/openvpn/update-resolv-conf
The first line enables the use of external scripts to handle the DNS implementation tasks. The up and down lines are there to implement DNS servers pushed by the VPN server when the connection goes up, and afterwards to undo it, when the connection goes down.
Ubuntu network management program
There is also the option of connecting through the GUI using the openvpn extension for the Gnome network manager plugin. But this is currently a bit tricky to set up. There is for example the incorrect assumption that all VPNs will be able to redirect Internet traffic, and older versions might not understand the .ovpn file format, requiring you to split up the certificate embedded in it into separate file. And you would likely have to dig into the options to ensure that a default Internet traffic route going through the VPN server is not always enabled by default, especially for servers where you only give access to some internal resources, and not the entire Internet. However the advantage of using the GUI component is that you can start/stop the connection from the desktop environment on Linux.
Connect to a VPN
A VPN (or Virtual Private Network ) is a way of connecting to a local network over the internet. For example, say you want to connect to the local network at your workplace while you’re on a business trip. You would find an internet connection somewhere (like at a hotel) and then connect to your workplace’s VPN. It would be as if you were directly connected to the network at work, but the actual network connection would be through the hotel’s internet connection. VPN connections are usually encrypted to prevent people from accessing the local network you’re connecting to without logging in.
There are a number of different types of VPN. You may have to install some extra software depending on what type of VPN you’re connecting to. Find out the connection details from whoever is in charge of the VPN and see which VPN client you need to use. Then, go to the software installer application and search for the NetworkManager package which works with your VPN (if there is one) and install it.
If there isn’t a NetworkManager package for your type of VPN, you will probably have to download and install some client software from the company that provides the VPN software. You’ll probably have to follow some different instructions to get that working.
To set up the VPN connection:
- Open the Activities overview and start typing Network .
- Click on Network to open the panel.
- At the bottom of the list on the left, click the + button to add a new connection.
- Choose VPN in the interface list.
- Choose which kind of VPN connection you have.
- Fill in the VPN connection details, then press Add once you are finished.
- When you have finished setting up the VPN, open the system menu from the right side of the top bar, click VPN off and select Connect . You may need to enter a password for the connection before it is established. Once the connection is made, you will see a lock shaped icon in the top bar.
- Hopefully you will successfully connect to the VPN. If not, you may need to double-check the VPN settings you entered. You can do this from the Network panel that you used to create the connection. Select the VPN connection from the list, then press the button to review the settings.
- To disconnect from the VPN, click the system menu on the top bar and click Turn Off under the name of your VPN connection.
More Information
- Wired networking — Use a wired internet connection and set a static IP address.
- Wireless networking — Connect to wireless networks, including hidden networks and networks created from phone tethering.
You can choose the displayed language by adding a language suffix to the web address so it ends with e.g. .html.en or .html.de.
If the web address has no language suffix, the preferred language specified in your web browser’s settings is used. For your convenience:
[ Change to English Language | Change to Browser’s Preferred Language ]
The material in this document is available under a free license, see Legal for details.
For information on contributing see the Ubuntu Documentation Team wiki page. To report errors in this documentation, file a bug.