Connection Wizard
The Connection Wizard walks you through the steps needed to set up a full WireGuard connection. You can use it to set up a new WireGuard interface from scratch (the Add Interface Wizard), or to add a new WireGuard endpoint to an existing interface (the Add Endpoint Wizard).
The wizard consists of five steps:
-
Type: You select the type of connection.
-
Identity: You select the identity of the two peers (and their hosts) to connect.
-
Transport: You specify the public IP address and port of one or both hosts.
-
Tunnel: You select the internal WireGuard IP addresses used by each peer; as well as specify any other network addresses that one peer can route to the other.
-
Review: You review the configuration settings that will be generated.
Within each step, fields shown on the left side of the page apply to the host or interface from which you starting the wizard (the “local” side of the connection), and fields shown on the right side of the page apply to the other side of the connection (the “remote” side of the connection). The specific fields and field names displayed will depend on the type of connection you choose in the first step.
See the WireGuard Connection Wizard blog post for several different examples of using the Connection Wizard.
Type
The first step of the wizard asks you to choose the type of connection to set up. Select the radio button corresponding to the type of connection you want:
Point-to-Point
A connection between two individual hosts in a simple point-to-point topology.
Site-to-Site
A connection between two networks in a basic site-to-site topology. Each network has host running WireGuard on it that can serve as a gateway to the other network; the wizard helps you configure the WireGuard interface on each host for this purpose.
Hub-and-Spoke, as Hub
A connection between a central hub and an individual spoke, in the classic hub-and-spoke topology, allowing each spoke to access other spokes. The hub and each spoke runs WireGuard; the wizard helps you set up the connection between the hub and one spoke, starting with the WireGuard interface on the hub.
Hub-and-Spoke, as Spoke
A connection between an individual spoke and a central hub, in the classic hub-and-spoke topology, allowing each spoke to access other spokes. The hub and each spoke runs WireGuard; the wizard helps you set up the connection between the hub and one spoke, starting with the WireGuard interface on the spoke.
Point-to-Site, as Site
A connection between a network and a single host not part of the network, using the point-to-site topology. The point runs WireGuard, and the network has a host in it that also runs WireGuard and serves as a gateway to the network for the point; the wizard helps you set up the connection between the point and the site gateway, starting with the WireGuard interface on the site gateway.
Point-to-Site, as Point
A connection between a single host and a network in which it is not a member, using the point-to-site topology. The point runs WireGuard, and the network has a host in it that also runs WireGuard and serves as a gateway to the network for the point; the wizard helps you set up the connection between the point and the site gateway, starting with the WireGuard interface on the point.
Point-to-Internet, as Internet
A connection between two hosts where one host provides Internet access to the other (essentially following the point-to-site topology, where the “site” is the Internet). Both hosts run WireGuard; the wizard helps you set up the connection between them, starting with the WireGuard interface on the Internet gateway.
Point-to-Internet, as Point
A connection between two hosts where one host provides Internet access to the other (essentially following the point-to-site topology, where the “site” is the Internet). Both hosts run WireGuard; the wizard helps you set up the connection between them, starting with the WireGuard interface on the point.
Identity
The second step of the wizard asks you to choose which two peer identities to connect: one on the left side of the page (the “local” side), to the other on the right (the “remote” side).
Local Peer
When adding an interface via the “Add Interface Wizard”, the wizard will ask you to select a peer on the left side of the page, to use as the peer identity for the WireGuard interface that you are adding. The public key of the peer you select will be used by WireGuard interfaces on other hosts to identity this interface.
You can either enter the name of a peer already registered with Pro Custodibus in the Peer field, or you can add a new peer by clicking the New button next to the Peer field. If you type part of a name in the Peer field, it will filter the displayed list of peers from which to choose. Use the up and down arrow keys to highlight a peer from the list, and use the tab or enter key to select the highlighted peer.
If you click the New button next to the Peer field, an Add Peer dialog will appear, allowing you to register a new peer identity with Pro Custodibus; you can then select this peer to use.
When adding an endpoint via the “Add Endpoint Wizard”, the wizard will not ask you to select a peer on the left side of the page — the peer identity for the local interface has already been set.
Local Host
The wizard will not ask you to select the host on the left side of the page — the local host has already been set (to the host from which you started the wizard).
Local Interface
When adding an interface via the “Add Interface Wizard”, after you select a local peer, you may set the name of the interface to add on the left side of the page. The wizard will auto-generate an interface name for you, such as “wg0”.
You can edit the auto-generated interface name in the Interface field if you prefer to name the interface differently. This name must be less than 16 characters, and must consist of only ASCII letters, numbers, dot (“.”), dash (“-”), and lodash (“_”) characters. It also must be unique among any other network interface names used by the host.
|
For macOS hosts, the interface name must either be “utun”, or “utun” and a number, like “utun0”, “utun1”, etc. For OpenBSD hosts, the interface name must either be “tun”, or “tun” and a number, like “tun0”, “tun1”, etc. |
This name can be changed later; but because it will be used by the host to identify the interface, changing it may break other things on the host that have been configured to interact with the interface.
You can also add a description for the interface, like “staging network for APAC”, in the Interface Description field. This description will be displayed alongside the interface name in various places in the UI. There are no length or character restrictions, and you can change it without consequence later.
When adding an endpoint via the “Add Endpoint Wizard”, you cannot change the interface name on the left side of the page — the name of the local interface has already been set.
Remote Peer
On the right side of the page, the wizard will ask you to select a peer to use as the peer identity for the other side of the connection.
You can either enter the name of a peer already registered with Pro Custodibus in the Peer field, or you can add a new peer by clicking the New button next to the Peer field. If you type part of a name in the Peer field, it will filter the displayed list of peers from which to choose. Use the up and down arrow keys to highlight a peer from the list, and use the tab or enter key to select the highlighted peer. Only peers not already used by another endpoint of the local interface will be listed.
If you click the New button next to the Peer field, an Add Peer dialog will appear, allowing you to register a new peer identity with Pro Custodibus; you can then select this peer to use.
Remote Host
On the right side of the page, the wizard will ask you to select the host for the other side of the connection. If the peer you selected above already is registered with a host and interface on Pro Custodibus, the wizard will automatically select that host and interface for you.
If the wizard did not automatically select a host, or if you want to use a different host, you can either enter the name of a host already registered with Pro Custodibus in the Host field, or you can add a new host by clicking the New button next to the Host field. If you type part of a name in the Host field, it will filter the displayed list of hosts from which to choose. Use the up and down arrow keys to highlight a host from the list, and use the tab or enter key to select the highlighted host.
If you click the New button next to the Host field, an Add Host dialog will appear, allowing you to register a new host with Pro Custodibus; you can then select this host to use.
Remote Interface
If the host you selected on the right side of the page does not already have a WireGuard interface registered for the peer you selected, you may set the name of the interface to register for it on the right side of the page. The wizard will auto-generate an interface name for you, such as “wg0”.
You can edit the auto-generated interface name in the Interface field if you prefer to name the interface differently. This name must be less than 16 characters, and must consist of only ASCII letters, numbers, dot (“.”), dash (“-”), and lodash (“_”) characters. It also must be unique among any other network interface names used by the host.
|
For macOS hosts, the interface name must either be “utun”, or “utun” and a number, like “utun0”, “utun1”, etc. For OpenBSD hosts, the interface name must either be “tun”, or “tun” and a number, like “tun0”, “tun1”, etc. |
This name can be changed later; but because it will be used by the host to identify the interface, changing it may break other things on the host that have been configured to interact with the interface.
You can also add a description for the interface, like “staging network for APAC”, in the Interface Description field. This description will be displayed alongside the interface name in various places in the UI. There are no length or character restrictions, and you can change it without consequence later.
Transport
The third step of the wizard asks you to specify the public IP address of one or both hosts (as well as its public WireGuard listen port). You must specify at least one side’s public IP address to enable WireGuard to set up a connection between the two hosts.
Local Host Has Public IP
If the local side of the connection (the host or interface for which you started the wizard) has a static public IP address, check the Host has a public IP address checkbox on the left side of the page, and enter the host’s IP address (or DNS name) into the Public IP Address or DNS Name field below the checkbox; and enter the listen port for the host’s WireGuard interface in the Listen Port field below its address.
| If you check the checkbox on the left side of the page, make sure you adjust any host- or network-based firewalls between the two hosts to allow the remote host to initiate new connections to the local host at the IP address and UDP port you’ve entered. |
Remote Host Has Public IP
If the remote side of the connection has a static public IP address, check the Host has a public IP address checkbox on the right side of the page, and enter the host’s IP address (or DNS name) into the Public IP Address or DNS Name field below the checkbox; and enter the listen port for the host’s WireGuard interface in the Listen Port field below its address.
| If you check the checkbox on the right side of the page, make sure you adjust any host- or network-based firewalls between the two hosts to allow the local host to initiate new connections to the remote host at the IP address and UDP port you’ve entered. |
IP Address
The IP address you enter into the Public IP Address or DNS Name field should be the host’s address from the perspective of the other side of the connection. For example, if the hosts on both sides of the connection are on the same LAN, you would enter the host’s LAN address. If the hosts are separated by the Internet, you would enter the Internet address of the host. If the host sits behind an Internet gateway, and you’re able to forward ports from the Internet gateway to the host, enter the Internet address of the gateway, and configure the gateway to forward the listen port to the host.
Listen Port
The port you enter into the Listen Port field should be the UDP port listening for new WireGuard connections from the perspective of the other side of the connection. When setting up a new WireGuard interface, the wizard will suggest a standard port number to use (usually “51820”).
The listen port cannot be in use by any other network service running on the host, including other WireGuard interfaces. If, when setting up a new interface, you select a port already in use by another WireGuard interface, the wizard will display a “must be unique on host” error message.
| When setting up an endpoint for an existing interface that does not already have a listen port configured, the wizard will display an “existing interface does not have a port set” warning message. You can proceed despite this warning message, but the new connection will not be available until the existing interface has been restarted to listen on the new port — which may interrupt existing connections currently using the interface. |
|
If you’ve set up port-forwarding on a gateway between the two hosts — and you’re forwarding from a source port that is different than the destination port — enter the source port in the Listen Port field; and if you’re adding a new interface, after you complete the wizard, edit the new interface to change its actual listen port to the destination port. (If the source and destination ports are the same, you don’t need to make any edits.) When you’re adding a new endpoint to an existing interface in this scenario (port forwarding with different source and destination ports), the wizard will display a “does not match existing interface port” warning message. As long as you’ve entered the correct source port, you can proceed with the wizard despite this warning. |
Tunnel
The fourth step of the wizard asks you to specify the internal WireGuard addresses of each side of the connection, as well as the network addresses that each side can route to the other through the encrypted WireGuard tunnel.
WireGuard Network Addresses
When setting up two brand new WireGuard interfaces via the “Add Interface Wizard”, the wizard will ask you to choose the network address space you want to use for the internal WireGuard IP addresses of the interfaces (you’ll then choose individual addresses below from within this larger space). Click the Generate button to use the wizard’s “Generate Network Address” dialog to generate an appropriate network address at random; or if you have already decided what address space to use, enter it directly.
Click the Add icon next to a network address to add another address; or (if you have multiple addresses) click the Remove icon to remove one of the addresses.
|
Most of the time, you will want to select either one IPv4 network address with a prefix length of To avoid conflicts with other IPv6 networks, the best practice for choosing IPv6 network address space is to generate at least the first 40 bits at random from the Alternatively, for WireGuard networks that don’t fit neatly into the site concept, choosing a random For IPv4, to minimize conflicts with other networks, you may want to select an IPv4 network address with a prefix length smaller than |
|
If the wizard displays a “address must fit prefix length” error message, it means the prefix length of a network address is too short (ie the netmask does not cover all the significant bits of the address). You must change either the network address or its prefix length to fit all of the address’s 1-bits. For example, the wizard will display the “address must fit prefix length” error message for a network address of |
If one WireGuard interface already exists (on either the local or the remote side), the wizard will present you the network address space which the existing interface appears to be using. You may add, remove, or edit the displayed networks; but your changes will apply only to the addresses of the new interface, not the existing one.
When using the “Add Endpoint Wizard” with two WireGuard interfaces which already exist, the wizard will show you the network address space the existing interfaces appear to be using — but you cannot make any changes to the interfaces' network space or addresses.
Local WireGuard Address
On the left side of the page, the wizard will display the internal IP addresses of the local WireGuard interface. These are the IP addresses the remote host (or connections forwarded by the remote host) should use in order to connect to any network services that the local host itself exposes through the WireGuard tunnel.
If the local WireGuard interface does not yet exist, the wizard will generate one unique IP address for the interface from the address space of each network address you selected with the WireGuard Network Addresses fields above. Click the Generate button next to an address to generate a new random IP address from within the corresponding network’s address space; or enter a different IP address of your choosing.
| If you enter an IP address that has already been assigned to a different WireGuard interface, the wizard will display an “already assigned” warning message. You can proceed despite this warning message — but beware that this can cause conflicts if you try to use both interfaces at the same time. Unless you have a specific reason to do so (eg replacing old hardware), don’t re-use WireGuard addresses. |
If the local WireGuard interface already exists, the wizard will display the existing IP addresses for the interface; you cannot change them in the wizard.
Route via Local Peer
On the left side of the page, the wizard will display the network addresses which the remote peer should route to the local peer. The wizard will generate one or more routable addresses based on the connection type and address space you’ve selected (which generally should include at least the WireGuard addresses of the local peer itself).
You can edit each routable address; or click the Add icon next to a routable address to add another; or click the Remove icon to remove one of the addresses.
|
If the local peer serves as a gateway to its local site from the WireGuard network (ie Point-to-Site as a Site, or Site-to-Site connection type), add the site’s network address space to this list (if not already added). For example, if the local peer’s host is on a LAN that uses the |
Remote WireGuard Address
On the right side of the page, the wizard will display the internal IP addresses of the remote WireGuard interface. These are the IP addresses the local host (or connections forwarded by the local host) should use in order to connect to any network services that the remote host itself exposes through the WireGuard tunnel.
If the remote WireGuard interface does not yet exist, the wizard will generate one unique IP address for the interface from the address space of each network address you selected with the WireGuard Network Addresses fields above. Click the Generate button next to an address to generate a new random IP address from within the corresponding network’s address space; or enter a different IP address of your choosing.
| If you enter an IP address that has already been assigned to a different WireGuard interface, the wizard will display an “already assigned” warning message. You can proceed despite this warning message — but beware that this can cause conflicts if you try to use both interfaces at the same time. Unless you have a specific reason to do so (eg replacing old hardware), don’t re-use WireGuard addresses. |
If the remote WireGuard interface already exists, the wizard will display the existing IP addresses for the interface; you cannot change them in the wizard.
Route via Remote Peer
On the right side of the page, the wizard will display the network addresses which the local peer should route to the remote peer. The wizard will generate one or more routable addresses based on the connection type and address space you’ve selected (which generally should include at least the WireGuard addresses of the remote peer itself).
You can edit each routable address; or click the Add icon next to a routable address to add another; or click the Remove icon to remove one of the addresses.
|
If the remote peer serves as a gateway to the remote site from the WireGuard network (ie Point-to-Site as a Point, or Site-to-Site connection type), add the site’s network address space to this list (if not already added). For example, if the remote peer’s host is on a LAN that uses the |
Options
Depending on the connection type you selected (and whether or not you’re creating a new interface), the wizard may present you with some additional options for the local host (left side) or the remote host (right side):
Turn on packet forwarding Linux kernel parameter
Check the Turn on packet forwarding Linux kernel parameter checkbox if a) the host is running Linux and b) it will not be running WireGuard in a container — and c) you haven’t already configured the host to turn on its net.ipv4.conf.all.forwarding or net.ipv6.conf.all.forwarding kernel parameters when it starts up.
|
On Linux and the BSDs (FreeBSD, OpenBSD, NetBSD, macOS, etc), you’d typically turn on these kernel parameters at boot via the host’s On Windows, you need to change the configuration of the “Routing and Remote Access” service to enable packet forwarding (see How can I enable packet forwarding on Windows?). |
Add packet masquerading iptables rule
Check the Add packet masquerading iptables rule checkbox if a) the host is running Linux and b) you’re using iptables to manage its firewall — and c) you want the host to masquerade connections forwarded from the WireGuard network with the host’s own LAN or public IP address.
|
If the host is serving as an Internet gateway for a “Point-to-Internet” connection type, you need to set up packet masquerading. However, if the host is serving as a site gateway for a “Point-to-Site” connection type, the best practice is to adjust the routing configuration at the site to enable the site to automatically route the WireGuard network via the host; and only if that is not feasible, configure the host’s firewall to masquerade packets forwarded from WireGuard. |
Use custom DNS settings when tunnel is up
Check the Use custom DNS settings when tunnel is up checkbox if you want the host to use different DNS resolvers or DNS search domains when its WireGuard connection is up (eg to use the private DNS servers and domain names for a private site when connected to it via WireGuard).
If you want the host to use different DNS resolvers, enter the IP address of the first resolver into the DNS Search field; click the Add icon next to the first DNS Search field to add more fields, and add the IP address of additional resolvers to each. If don’t want to change the host’s DNS resolvers, simply leave the DNS Search field blank.
If you want the host to use different DNS search domains, enter the first domain name into the Search Domain field; click the Add icon next to the first Search Domain field to add more fields, and add an additional domain name to each. If don’t want to change the host’s DNS search domains, simply leave the Search Domain field blank.
|
If the host uses systemd-resolved as its internal DNS resolver, you may want to instead use |
Use Quad9 DNS servers when tunnel is up
Select the Use Quad9 DNS servers when tunnel is up radio button if the host a) normally uses its ISP’s (Internet Service Provider’s) DNS resolvers (usually acquired via DHCP) and b) you want it to instead use the Quad9 DNS resolvers when the WireGuard connection is up (eg for privacy).
Review
The final step of the wizard shows you the WireGuard configuration settings it will generate. On the left side of the page, it will display the configuration for the host or interface from which you started the wizard (the “local” side of the connection); and on the right side of the page, it will display the configuration for the other side of the connection (the configuration for the “remote” interface).
The parts of the configuration highlighted with a green background indicate the new sections of the configuration. If an interface is new, its entire configuration will be displayed with a green background; if only the endpoint is new, only the section for the new endpoint will be highlighted with a green background.
|
If an existing interface has other existing endpoints, they will not be shown on this page; only the new endpoint will be displayed. |
Click the Apply button to submit the form and queue the creation of the interfaces or endpoints.
Complete Set-Up
After you click the Apply button in the wizard, the next time the Pro Custodibus agent on each host pings the Pro Custodibus servers, the agent will receive the configuration for the new interface or endpoint, and apply it to the host. The “Complete Set-Up” page will display links to the queued changes for both sides of the connection.
If you have not yet set up the agent on one side of the connection or the other, choose one of the following options:
-
Click the Set Up Agent icon (“gear” icon) for that side of the connection, and follow the steps to set up an agent on the host (you can also do this at any time later by following the steps from the Set Up a Host documentation).
-
Click the Download Interface Config icon (“QR code” icon) for that side of the connection, and download the WireGuard configuration file (or QR code) for the interface (you can also do this at any time later by following the steps from the Download Configuration documentation). Manually apply the downloaded configuration to the host (and then mark its queued changes as applied via the steps described in the Change Queue documentation).
From the “Complete Set-Up” page, you can also click the Add Another Endpoint icon (“plus” icon) for one side of the connection or the other to add another new connection to some other WireGuard peer, starting with the WireGuard interface on that side of the connection. You can also do this at any time later by following the steps from the Add an Endpoint documentation.
Also see the “Set Up Pro Custodibus Agent on Server” and “Download WireGuard Config and Apply to Second Workstation” sections of the WireGuard Connection Wizard blog post for two different examples of completing the set-up process.