WireGuard Remote Access VPN Configuration Example¶

This recipe covers configuring a basic WireGuard remote access style VPN tunnel.

Note

Though WireGuard does not have a concept of “Client” and “Server” per se, in this style of deployment the firewall cannot initiate connections to remote peers. In this way the firewall acts like a “Server” and may be referred to as such in this documentation. Remote peers may also be referred to as “clients”.

Required Information¶

The following basic information must be determined before starting the VPN configuration.

Item	Value
Design	Remote access, one tunnel+many peers
Firewall WAN	`198.51.100.6`
Listen Port	`51820`
Tunnel Subnet	`10.6.210.0/24`
Tunnel Address	`10.6.210.1/24`
Peer Addresses	`10.6.210.2` - `10.6.210.254`
Peer Endpoints	Dynamic

Generating Keys¶

WireGuard requires public/private key pairs for each peer, including this firewall.

Warning

Keys cannot be reused between clients, as WireGuard requires unique keys to identify clients and where to send their traffic.

Tunnel Keys¶

To generate keys for the firewall itself, click the Generate button when configuring a tunnel. The GUI will populate the private and public key fields automatically.

The peers will need the public key for their configuration.

Peer Keys¶

Each peer will need its own public/private key pair. The private key will be needed on the peer client software while the public key will be needed on the firewall itself for the peer definition.

These keys can be generated by the clients themselves, or via command line on a system which has the WireGuard utilities installed. This includes the firewall itself; these commands may be run from a console or SSH shell or from Diagnostics > Command Prompt.

From a command line, execute the following:

$ wg genkey | tee privatekey | wg pubkey > publickey

This command outputs files named privatekey and publickey which respectively contain a private key and its associated public key. This key pair can be used for a WireGuard peer.

To view the keys, inspect the contents of the files:

$ cat privatekey
WGpL3/ejM5L9ngLoAtXkSP1QTNp4eSD34Zh6/Jfni1Q=
$ cat publickey
b9FjbupGC7fomO5U4jL5Irt1ZV5rq4c+utGKj53HXgU=

Repeat the commands as needed as many times as is necessary for the number of peers required by this tunnel. Note the keys in a secure place.

Tip

Change the commands to output files named for their associated peer, then store the resulting files in a secure location.

Alternately, the keys can be output in one command without storing them persistently. This behavior is not be supported on all platforms, but is supported on the firewall itself.

$ wg genkey | tee /dev/stderr | wg pubkey
4BSH81zC3/OWl25XrzqWy7WnAiARXySHd+K+KFxNrWU=
rzWOC0zH9v2zF6r92uCbjs7JOmhqy8N+cUdA+GCynSM=

Tunnel Configuration¶

Now it’s time to create the WireGuard tunnel.

Navigate to VPN > WireGuard > Tunnels
Click Add Tunnel
Fill in the options using the information determined earlier:

Enable:

Checked

Description:

Remote Access

Listen Port:

51820

Interface Keys:

Click Generate to create a new set of keys.

Interface Addresses:

10.6.210.1/24
Click Save

Peer Configuration¶

Peers can be added when editing a tunnel. To edit a tunnel:

Navigate to VPN > WireGuard > Peers
Click Add Peer
Fill in the options using the information determined earlier:

Enable:

Checked

Tunnel:

tun_wg<num> (Remote Access)

Description:

The name of this client (e.g. The name of a person, device, username, or other uniquely identifying information.)

Dynamic Endpoint:

Checked

Keep Alive:

Typically left blank, but may be filled in if clients have problems traversing certain firewalls.

Public Key:

The public key for this peer. Obtained from the key generation process earlier, or from the peer itself if it was generated by client software directly.

Pre-Shared Key:

Not used in this example, but for additional security this pre-shared key can be generated and copied to the peer. Must match on the client and server.

Allowed IPs:

The tunnel IP address for this peer, from the list determined above, with a /32 CIDR mask. For example, the first peer will be 10.6.210.2/32, the second will be 10.6.210.3/32, and so on.
Click Save Peer
Repeat the steps to add additional peers as needed.

Firewall Rules¶

First add a rule to pass external WireGuard traffic on the WAN:

Navigate to Firewall > Rules, WAN tab
Click Add to add a new rule to the top of the list
Use the following settings:

Action:

Pass

Interface:

WAN

Protocol:

UDP

Source:

any

Destination:

WAN Address

Destination Port Range:

(other), 51820

Description:

Pass traffic to WireGuard
Click Save
Click Apply Changes

Next, add a rule to pass traffic inside the WireGuard tunnel:

Navigate to Firewall > Rules, WireGuard tab
Click Add to add a new rule to the top of the list
Use the following settings:

Action:

Pass

Interface:

WireGuard

Protocol:

Any

Source:

any

Destination:

any

Description:

Pass VPN traffic from WireGuard peers
Click Save
Click Apply Changes

Client Configuration¶

Client configuration varies by platform, see WireGuard documentation for details. This section covers a basic configuration.

This is an example configuration from a WireGuard client for a split-tunnel configuration:

[Interface]
PrivateKey = WGpL3/ejM5L9ngLoAtXkSP1QTNp4eSD34Zh6/Jfni1Q=
ListenPort = 51820
Address = 10.6.210.2/24

[Peer]
PublicKey = PUVBJ+zuz/0mRPEB4tIaVbet5NzVwdWMX7crGx+/wDs=
AllowedIPs = 10.6.210.1/32, 10.6.0.0/24
Endpoint = 198.51.100.6:51820

This is an example configuration from a WireGuard client for a full-tunnel configuration:

[Interface]
PrivateKey = WGpL3/ejM5L9ngLoAtXkSP1QTNp4eSD34Zh6/Jfni1Q=
ListenPort = 51820
DNS = 10.6.210.1, pfSense.home.arpa
Address = 10.6.210.2/24

[Peer]
PublicKey = PUVBJ+zuz/0mRPEB4tIaVbet5NzVwdWMX7crGx+/wDs=
AllowedIPs = 0.0.0.0/0
Endpoint = 198.51.100.6:51820

The fields in that file are as follows:

Interface:

Settings for this client.

PrivateKey:

The private key for this peer. Obtained from the key generation process earlier, or from the peer itself if it was generated by client software directly.

ListenPort:

A static port to listen on, or omit the line to use a random port instead.

DNS:

The DNS server(s) and search domain that should be used by the system when the tunnel is enabled.

Address:

The tunnel address for this client. Not supported on all platforms, as some require configuring the address using command-line utilities. However, clients on Windows and Android, for example, support this directive.

This should use the same CIDR mask as the Tunnel address. In this example, the first peer is 10.6.210.2/24.

Peer:

Configuration for the firewall end of the tunnel.

PublicKey:: The public key from the Tunnel configuration on the firewall.
AllowedIPs:: The Tunnel address, and any additional networks which should be routed across the VPN in a comma-separated list. This could be a LAN subnet (e.g. 10.6.0.0/24) or use 0.0.0.0/0 to route all traffic, including Internet traffic, across the tunnel.
Dynamnic Endpoint:: Unchecked
Endpoint:: The firewall WAN IP address and WireGuard Listen Port

Note

This only covers the basics, there are numerous other fields which can be used to control client behavior plus additional client options which vary by platform. For additional details, see the WireGuard documentation and the documentation for the WireGuard software used by a peer.

Transfer the resulting client configuration file to the peer in a secure manner. Methods vary by platform and client software.

Finish Up¶

After configuring the client and activating the VPN, the client should be able to pass traffic to the networks listed in the AllowedIPs list in its configuration.