Connecting Multiple Vultr Zones With N2N

Updated on November 21, 2023
Connecting Multiple Vultr Zones With N2N header image

N2N is an open source layer 2/3 VPN application. Unlike many other VPN programs, N2N can connect computers which are located behind a NAT router. This offers a huge advantage to connect to a cloud environment without having to rely on special protocols such as the ESP protocol (used by ipsec). To achieve this connection, N2N uses a supernode which can route the information between NAT'ed nodes. This VPN connection can be used to connect multiple Vultr instances across regions together.

Prerequisites

In this example we will be using three nodes in multiple zones:

  • Paris
  • Miami
  • Sydney

Installation of the software

The following commands will be executed on each instance.

Start by installing the build-essential from the repo and also libssl-dev, as we will be building from the newest source code.

apt-get install -y build-essential libssl-dev

Next, download the source code from github.

cd /tmp
git clone https://github.com/ntop/n2n.git

Compile all binaries.

cd n2n 
make
make install

The make install command will have created the supernode and edge binaries in the /usr/sbin directory. Finish by cleaning up the files.

rm -rf /tmp/n2n

Installation - Node Paris

The first node will be our so called supernode. This supernode will start the supernode service that will listen on UDP port 1200. By default the N2N application doesn't create a service file. So we will need to provide our own.

Create the 'n2n_supernode' service file:

nano /etc/systemd/system/n2n_supernode.service

Add the following content:

[Unit]
Description=n2n supernode
Wants=network-online.target
After=network-online.target

[Service]
ExecStart=/usr/sbin/supernode -l 1200

[Install]
WantedBy=multi-user.target

The '-l' directive defines the UDP port 1200. This is the port on which the supernode will listen. To ensure that everything is working, start the supernode service:

systemctl start n2n_supernode

Check the status of the supernode.

systemctl status n2n_supernode

This will show a status similar to the following.

● n2n_supernode.service - n2n supernode
   Loaded: loaded (/etc/systemd/system/n2n_supernode.service; disabled; vendor prese
   Active: active (running) since Wed 2018-08-15 17:07:46 UTC; 5s ago
 Main PID: 4711 (supernode)
    Tasks: 1
   Memory: 80.0K
      CPU: 1ms
   CGroup: /system.slice/n2n_supernode.service
           └─4711 /usr/sbin/supernode -l 1200

Next we will create the edge service. This edge service will claim a private IP for communication between the other edges in other Vultr zones.

As with the supernode service, this will also need its own service file.

nano /etc/systemd/system/n2n_edge.service

Add the following content:

[Unit]
Description=n2n edge
Wants=network-online.target
After=network-online.target n2n_supernode.service

[Service]
ExecStart=/usr/sbin/edge -l localhost:1200 -c Vultr -a 192.168.1.1 -k mypassword -f

[Install]
WantedBy=multi-user.target

In this service file we defined the following command line options:

  • -l localhost:1200: This will connect to localhost on UDP port 1200.
  • -c Vultr: This is the community the edge will be joining. All edges within the same community appear on the same LAN (layer 2 network segment). Edges that are not in the same community will not communicate with each other.
  • -a 192.168.1.1: The IP assigned to this interface. This is the N2N virtual LAN IP address being claimed.
  • -k mypassword: The password used for each edge. All edges communicating must use the same key and community name.
  • -f: Disables daemon mode and causes edge to run in the foreground. This is needed for the service file, otherwise systemctl will not start the service.

To ensure that everything is working, start the service.

systemctl start n2n_edge

Then, query the service status.

systemctl status n2n_edge	

The output will be similar to the following.

● n2n_edge.service - n2n edge
   Loaded: loaded (/etc/systemd/system/n2n_edge.service; disabled; vendor preset: en
   Active: active (running) since Wed 2018-08-15 17:10:46 UTC; 3s ago
 Main PID: 4776 (edge)
    Tasks: 1
   Memory: 396.0K
      CPU: 8ms
   CGroup: /system.slice/n2n_edge.service
           └─4776 /usr/sbin/edge -l localhost:1200 -c Vultr -a 192.168.1.1 -k mypass

If we check 'ifconfig', you will see the N2N virtual IP being claimed by the edge0 interface.

ifconfig

The output will be similar to the following.

edge0     Link encap:Ethernet  HWaddr 42:14:55:64:7d:21
          inet addr:192.168.1.1  Bcast:192.168.1.255  Mask:255.255.255.0
          inet6 addr: fe80::4014:55ff:fe64:7d21/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1400  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:8 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:0 (0.0 B)  TX bytes:648 (648.0 B)

Once this is done, enable and create the firewall rules. Make sure to replace the node_miami_ip and node_sydney_ip text with the public IP of the Sydney and Miami instance. (We will use these later on).

ufw allow 22/tcp
ufw allow from node_miami_ip to any port 1200
ufw allow from node_sydney_ip to any port 1200
ufw enable

The last thing to do with this node is to enable both services at boot.

systemctl enable n2n_supernode.service
systemctl enable n2n_edge.service

Installation - Node Miami

The Miami node will connect to the super node which is currently running in the Paris zone. To achieve this we only need to create a service file for the edge application.

Start by creating an edge service file.

nano /etc/systemd/system/n2n_edge.service

Add the following content.

[Unit]
Description=n2n edge
Wants=network-online.target
After=network-online.target

[Service]
ExecStart=/usr/sbin/edge -l node_paris_ip:1200 -c Vultr -a 192.168.1.2 -k mypassword -f

[Install]
WantedBy=multi-user.target

Note: Replace the node_paris_ip with the public IP of the instance running in Paris

This will connect to the node in Paris on UDP port 1200, join community 'Vultr', claim IP 192.168.1.2 and authenticate with 'mypassword'.

Next, start the service.

systemctl start n2n_edge

Check the status for an indication that the service has started correctly and is running.

systemctl status n2n_edge	

Next, ensure that the edge0 IP gets claimed.

ifconfig

It will show the 192.168.1.2 IP address.

edge0     Link encap:Ethernet  HWaddr 42:14:55:64:7d:21
          inet addr:192.168.1.2  Bcast:192.168.1.255  Mask:255.255.255.0
          inet6 addr: fe80::4014:55ff:fe64:7d21/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1400  Metric:1
          RX packets:0 errors:0 dropped:0 overruns:0 frame:0
          TX packets:8 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:0 (0.0 B)  TX bytes:648 (648.0 B)

The next thing to do is to enable the service at boot.

systemctl enable n2n_edge.service

Optionally, enable the firewall and add the SSH rules.

ufw allow 22/tcp
ufw enable

We will now be able to ping both edges running in our instances.

In Paris, ping the Vultr instance in Miami

ping 192.168.1.2

In Miami, ping the edge in Paris

ping 192.168.1.1

Installation - Node Sydney

Finally, we will add our last continent to the mix: Australia. Start again by creating an edge service, this edge service will also connect to the previous configured supernode in Paris.

nano /etc/systemd/system/n2n_edge.service

Add the following content.

[Unit]
Description=n2n edge
Wants=network-online.target
After=network-online.target

[Service]
ExecStart=/usr/sbin/edge -l node_paris_ip:1200 -c Vultr -a 192.168.1.3 -k mypassword -f

[Install]
WantedBy=multi-user.target

Note: Replace the node_paris_ip with the public IP of the instance running in Paris.

This will connect to the node in Paris on UDP port 1200, join community 'Vultr', claim IP 192.168.1.3 and authenticate with 'mypassword'.

systemctl start n2n_edge

Check the status to ensure the service is started.

systemctl status n2n_edge	

Make sure that the edge0 IP gets claimed.

edge0     Link encap:Ethernet  HWaddr 46:56:b0:e9:8f:8a
    	  inet addr:192.168.1.3  Bcast:192.168.1.255  Mask:255.255.255.0
      	inet6 addr: fe80::4456:b0ff:fee9:8f8a/64 Scope:Link
      	UP BROADCAST RUNNING MULTICAST  MTU:1400  Metric:1
      	RX packets:0 errors:0 dropped:0 overruns:0 frame:0
      	TX packets:8 errors:0 dropped:0 overruns:0 carrier:0
      	collisions:0 txqueuelen:1000
      	RX bytes:0 (0.0 B)  TX bytes:648 (648.0 B)

Again, enable this service at boot.

systemctl enable n2n_edge.service

Optionally, enable the firewall and add the SSH rules.

ufw allow 22/tcp
ufw enable

We will now be able to ping each Vultr instance from each node.

ping 192.168.1.1
ping 192.168.1.2
ping 192.168.1.3

If you want to test the connection between each node edge, enable the firewall rules on the instances of Miami and Paris. This will allow communication between edges.

In Miami, add the following rules. (Make sure to replace the node_paris_ip and node_sydney_ip text with the public IPs of the Sydney and Paris instances.)

ufw allow from node_paris_ip to any port 1200
ufw allow from node_sydney_ip to any port 1200

In Sydney, add the following rules.

ufw allow from node_paris_ip to any port 1200
ufw allow from node_miami_ip to any port 1200

Now you can shutdown or reboot the supernode. Network connections will continue to exist. Only new edges will suffer connectivity issues while the supernode service is down.

Conclusion

We have successfully configured a VPN connection between multiple zones. This should offer quite a lot of new possibilities for high availability scenarios to our newly configured environment.