OpenVPN - How to Set Up a Slackware Server and a Slackware Client

OpenVPN is an open source software application that implements virtual private network (VPN) techniques for creating secure point-to-point or site-to-site connections in routed or bridged configurations and remote access facilities. It uses a custom security protocol that utilizes SSL/TLS for key exchange. It is capable of traversing network address translators (NATs) and firewalls. It was written by James Yonan and is published under the GNU General Public License (GPL).

OpenVPN allows peers to authenticate each other using a pre-shared secret key, certificates, or username/password. When used in a multiclient-server configuration , it allows the server to release an authentication certificate for every client, using signature and Certificate authority. It uses the OpenSSL encryption library extensively, as well as the SSLv3/TLSv1 protocol, and contains many security and control features.

The objective of this article is to serve as a tutorial for the readers to set up a basic but functional Slackware VPN Server and Client over the Internet.

The emphasis is to provide a reliable method that can be easily followed to set-up OpenVPN on Slackware Servers and Clients. Nevertheless the process is still not free from pitfalls and require some attention and determination.

This article comprises of a selection of other similar tutorials found on Internet particularly (2) and (3) and the documents contained in the downloaded source files. However these are reformatted to satisfy the objective.

Openvpn is already installed on Slackware if a default installation was followed. If this was not the case, then the package is available from the “n” directory of the Slackware DVD. Refer to other Slackware specific documents on how to go about this installation.

If you want to confirm that Openvpn is indeed installed, you can check it by listing the /var/log/packages/ directory:

# ls /var/log/packages/openvpn*

Server and a Client computers would be needed. They would have to be connected to the Internet on two different Routers and different Network Routes. For the purpose of this tutorial, specific details are defined in order to enhance the readability. Of course, you will probably have a different addresses, so you will need to amend accordingly.

A URL is normally used to address the Server. This is not mandatory and instead you may use only the Internet IP. However it is recommended to use a URL to access the Server from the Internet, especially if it is connected to a dynamic IP, which is typical for domestic Internet connections. The author is using noip2(4) as it is free upon subscription. Noip2 is available from http://slackbuilds.org.

hostname: server1
IP: 192.168.200.195/255.255.255.0
URL:  servervpn.no-ip.org
Network Interface: eth0

hostname: client1
IP: 192.168.1.101/255.255.255.0
Network Interface: wlan0

You will need to have administrator rights to set up the Openvpn. This applies to both the Server and the Client. For simplicity, in this tutorial, it will be assumed that all actions will be performed by the root user. Naturally advanced users might be more discerning.

The availability of two Routers might be challenging. Consider that interactive sessions on both the Server and Client will be needed before the VPN is set up. If the Client is equipped with a WiFi interface there might be some easy solutions that may be considered:

1. Use a 3G smart phone's “Portable Wi-Fi Hot Spot” facility to connect the Client as the VPN Client. As 3G bandwidth is expensive you may want to minimise traffic. For example, you might want to switch off services that are not absolutely essential during the course of this exercise, such as ntpd, dropbox and tor.
2. Connect the Client to another WiFi available in the vicinity of the Server. Some lucky people live in areas were benevolent neighbours provide them with openly accessible Internet WiFi. It is recommended to request permission before taking up this solution. In case that no such open service exists, you may find it appropriate to request a temporary password from a friendly neighbour for the private encrypted WiFi service.
3. Nowadays, many governmental premises, such as libraries and Local Councils provide free WiFi service. Other places such as fast food outlets, pubs, cafés, etc. also provide free WiFi from their location to their valuable customers. You may access the Server via an available service such as SSH from a WiFi equipped Client. If this option is chosen for this solution, be aware that the Client may have to pass through some firewalls. Besides the VPN connection might be a breach of the terms of conditions that should be accepted before using the WiFi service.

The PKI may be created on any computer with a VPN installation, but it is probably more sensible to be done on both the Server and the Client as both would need it. An easy way to build the PKI is to use the easy-rsa scripts. These may be downloaded like this:

# cd
# git clone git://github.com/OpenVPN/easy-rsa

and then archive it for future purposes:

# tar cvf easy-rsa.tar easy-rsa

Follow these steps on the Server to create the needed keys and certificates:

# cd easy-rsa/easyrsa3

Create the PKI and the CA:

# ./easyrsa init-pki
# ./easyrsa build-ca

Enter a PEM pass phrase, reverify it and then enter a name for the server. In this article I am using the hostnames for clarity (in this case: server1), but you may choose any name.

Then generate the request:

# ./easyrsa gen-req server1

You will be prompted for another PEM pass phrase to reverify it and to confirm that the name of the entity is indeed server1. Now you may proceed to sign this request:

# ./easyrsa sign-req server server1

Confirm the request by entering “yes”, then enter original ca PEM passphrase.

Now create two additional key files:

# cd /etc/openvpn/certs/
# openssl dhparam -out dh2048.pem 2048
# cd /etc/openvpn/keys/
# /usr/sbin/openvpn --genkey --secret ta.key 

Copy the following files generated by the easy-rsa scripts to their respective directories in the /etc/openvpn/ directory:

# cp /root/easy-rsa/2.0/keys/{ca.crt,server1.crt,dh2048.pem} \
> /etc/openvpn/certs/
# cp /root/easy-rsa/2.0/keys/{server1.key,ta.key} \
> /etc/openvpn/keys/

Restrict the permission of /etc/openvpn/keys/

# chmod 700 /etc/openvpn/keys

On the Server machine copy the sample server.conf onto the openvpn's configuration directory.

# cp /usr/doc/openvpn-2.2.2/sample-config-files/server.conf \
> /etc/openvpn/

Edit the following lines of /etc/openvpn/server.conf

From these lines:

ca ca.crt
cert server.crt
key server.key  # This file should be kept secret


dh dh1024.pem
;tls-auth ta.key 0 # This file is secret

;user nobody 
;group nobody

To

ca /etc/openvpn/certs/ca.crt 
cert /etc/openvpn/certs/server1.crt 
key /etc/openvpn/keys/server1.key #This file should be kept secret

dh /etc/openvpn/certs/dh2048.pem

tls-auth /etc/openvpn/keys/ta.key 0 # This file is secret

user nobody 
group nobody

Note that comments in server.conf may be either start with # or ; In order to help you with entering parameters, the former are used to comment out text while the latter are for commented out configuration lines.

Copy the rc.openvpn-server listed hereunder and place under /etc/rc.d/

#!/bin/sh 
# 
# /etc/rc.d/rc.openvpn-server 
# 
# Start/stop/restart the openvpn server. 
# 
# To make OpenVPN start automatically at boot, make this 
# file executable:  chmod 755 /etc/rc.d/rc.openvpn-server 
# 

ovpn_start() { 
  if [ -x /usr/sbin/openvpn -a -r /etc/openvpn/server.conf ]; then 
    echo "Starting OpenVPN:  /usr/sbin/openvpn server.conf" 
    /usr/sbin/openvpn /etc/openvpn/server.conf & 
  fi 
} 

ovpn_stop() { 
  killall openvpn 
} 

ovpn_restart() { 
  ovpn_stop 
  sleep 2 
  ovpn_start 
} 

case "$1" in 
'start') 
  ovpn_start 
  ;; 
'stop') 
  ovpn_stop 
  ;; 
'restart') 
  ovpn_restart 
  ;; 
*) 
  echo "Usage: $0 {start|stop|restart}" 
esac

Then give it executable permissions:

# chmod 755 /etc/rc.d/rc.openvpn-server

You will need to forward traffic from the port you have chosen for Openvpn to be routed to the Server. To accomplish this you will need to provide your Server with a fixed IP and you will need to configure your router. You may use netconfig, wicd or network-manager to set the fixed IP on Slackware. Then you also need to consult the documentation provided with your router to set up the selected IP address reserved for the Server, and the port forwarding. For our default Openvpn set up, the UDP Port would be 1194.

In case if you have misplaced such documentation, you may search on the Internet on how this may be achieved. A good place to start is http://portforward.com/. Astute readers are warned that typical port scanning tools like nmap might not be able to detect your open ports at this stage.

On the client machine perform the following instructions to set it up.

# cp /usr/doc/openvpn-2.2.2/sample-config-files/client.conf \
> /etc/openvpn/

Edit the following lines of /etc/openvpn/client.conf

remote my-server-1 1194

;user nobody 
;group nobody 

ca ca.crt 
cert client.crt 
key client.key 

;tls-auth ta.key 1

to the following lines:

remote servervpn.no-ip.org 1194

user nobody 
group nobody 

ca /etc/openvpn/certs/ca.crt 
cert /etc/openvpn/certs/client1.crt 
key /etc/openvpn/keys/client1.key 

tls-auth /etc/openvpn/keys/ta.key 1

Note that comments in client.conf may be either # or ; The former are used to comment out text while the latter are for commented out configuration lines. This should help you a lot in the configuration process.

You will need the following files that were generated by the easy-rsa scripts:

ca.crt
client1.crt
client1.key
ta.key

Place these files as indicated in client.conf. So ca.crt and client1.crt go under /etc/openvpn/certs/ while client1.key and ta.key go under /etc/openvpn/keys/

Copy the following rc.openvpn-client listed hereunder and place under /etc/rc.d/

#!/bin/sh 
# 
# /etc/rc.d/rc.openvpn-client 
# 
# Start/stop/restart the openvpn client. 
# 

ovpn_start() { 
  if [ -x /usr/sbin/openvpn -a -r /etc/openvpn/client.conf ]; then 
    echo "Starting OpenVPN:  /usr/sbin/openvpn client.conf" 
    /usr/sbin/openvpn /etc/openvpn/client.conf & 
  fi 
} 

ovpn_stop() { 
  killall openvpn 
} 

ovpn_restart() { 
  ovpn_stop 
  sleep 2 
  ovpn_start 
} 
 
case "$1" in 
'start') 
  ovpn_start 
  ;; 
'stop') 
  ovpn_stop 
  ;; 
'restart') 
  ovpn_restart 
  ;; 
*) 
  echo "Usage: $0 {start|stop|restart}" 
esac

Then give it executable permissions:

# chmod +x /etc/rc.d/rc.openvpn-client

On the Server:

# /etc/rc.d/rc.openvpn-server start

On the Client:

# /etc/rc.d/rc.openvpn-client start

On both you should see a new network interface called tun0. On the Server, I obtained the following:

# ifconfig tun0 
tun0: flags=4305<UP,POINTOPOINT,RUNNING,NOARP,MULTICAST>  mtu 1500 
        inet 10.8.0.1  netmask 255.255.255.255  destination 10.8.0.2 
        unspec 00-00-00-00-00-00-00-00-00-00-00-00-00-00-00-00  txqueuelen 100  (UNSPEC) 
        RX packets 0  bytes 0 (0.0 B) 
        RX errors 0  dropped 0  overruns 0  frame 0 
        TX packets 0  bytes 0 (0.0 B) 
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0 

Similarly on the Client:

# ifconfig tun0 
tun0: flags=4305<UP,POINTOPOINT,RUNNING,NOARP,MULTICAST>  mtu 1500 
        inet 10.8.0.6  netmask 255.255.255.255  destination 10.8.0.5 
        unspec 00-00-00-00-00-00-00-00-00-00-00-00-00-00-00-00  txqueuelen 100  (UNSPEC) 
        RX packets 0  bytes 0 (0.0 B) 
        RX errors 0  dropped 0  overruns 0  frame 0 
        TX packets 0  bytes 0 (0.0 B) 
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

Naturally you can ping the Server from Client (or vice versa):

For example, from the Client:

# ping -c 3 10.8.0.1
PING 10.8.0.1 (10.8.0.1) 56(84) bytes of data. 
64 bytes from 10.8.0.1: icmp_req=1 ttl=64 time=2888 ms 
64 bytes from 10.8.0.1: icmp_req=2 ttl=64 time=1997 ms 
64 bytes from 10.8.0.1: icmp_req=3 ttl=64 time=1324 ms 

--- 10.8.0.1 ping statistics --- 
3 packets transmitted, 3 received, 0% packet loss, time 1999ms 
rtt min/avg/max/mdev = 1324.475/2070.293/2888.429/640.527 ms, pipe 3

To start the openvpn service automatically on boot-up from the Server, include these lines in /etc/rc.d/rc.local

# Start the OpenVPN Server
if [ -x /etc/rc.d/rc.openvpn-server ]; then 
  /etc/rc.d/rc.openvpn-server start 
fi

Up to now we have created a tunnel device on both the Server and the Client called tun0 which is visible only to these two machines. However more work is needed to route the Client's connection via tun0 and then to the WAN that is connected to the Server.

Enable IP forwarding:

# chmod +x /etc/rc.d/rc.ip_forward
# /etc/rc.d/rc.ip_forward start

IP forwarding is now enabled and will be enabled also after you reboot.

Make a directory called ccd in /etc/openvpn

# mkdir /etc/openvpn/ccd/

Create a file with the same name of the client (in this case client1) and enter the following line in /etc/openvpn/ccd/client1

iroute 192.168.1.0 255.255.255.0

Replace 192.168.1.0 255.255.255.0 by the Network Route of your Client.

Similarly edit /etc/openvpn/server.conf with the following lines:

push "route 192.168.200.0 255.255.255.0"

client-config-dir ccd 
route 192.168.1.0 255.255.255.0 

push "redirect-gateway def1 bypass-dhcp"

push "dhcp-option DNS 208.67.222.222" 
push "dhcp-option DNS 208.67.220.220"

Naturally replace 192.168.200.0 255.255.255.0 with the Server's Network Route, and 192.168.1.0 255.255.255.0 with the Client's Network Route.

208.67.222.222 and 208.67.220.220 are the OpenDNS IP addresses. Up to now the DNS push configuration has not been successful. You can either use the original Client DNS servers or else you may edit /etc/resolv.conf manually. However it is still worth trying to use the DNS servers listed by the Client, I find that they are generally still be available from the Server's Network, so you would not need to do anything.

Next you will have to configure some iptables NAT forwarding on the Server (only). You can do this by first flushing the iptables:

# iptables -F

And then:

# iptables -t nat -A POSTROUTING -s 10.8.0.0/24 -o eth0 -j MASQUERADE

On Slackware, such a line may be included in /etc/rc.d/rc.firewall and /etc/rc.d/rc.inet2 will run it each time you reboot the Server if the former has executable permissions. You do not have to include anything in /etc/rc.d/rc.local.

The exact lines which you need to include depend on whether you already entered your own iptables filter chains and rules, but I will assume that that this is not the case.

As already explained, as a minimum you only need to enter the following lines in /etc/rc.d/rc.firewall

#!/bin/sh
iptables -t nat -A POSTROUTING -s 10.8.0.0/24 -o eth0 -j MASQUERADE

If on the other hand you would like a better firewall and you are at least moderately confident with iptables, I propose the following script to be included in your /etc/rc.d/rc.firewall. The comments in the script should help you understand the impact they will have on the Server.

#!/bin/sh 
# Start/stop/restart the firewall 
# A VPN server iptable filter rules 
 
IPT=/usr/sbin/iptables # This will provide some portability 

firewall_start() { 

  # Flush the iptables 
  $IPT -F 

  # Policies 
  $IPT -P OUTPUT ACCEPT 
  $IPT -P INPUT DROP 
  $IPT -P FORWARD DROP 
  $IPT -N SERVICES # Services is a custom chain 

  # Allowed inputs 
  #$IPT -A INPUT -i lo -j ACCEPT # Uncomment if the Server is a desktop 
  $IPT -A INPUT -m conntrack --ctstate ESTABLISHED,RELATED -j ACCEPT # Allow responses
  $IPT -A INPUT -j SERVICES # Append the services chain to the input 

  # Allowed forwards 
  $IPT -A FORWARD -i eth0 -o tun0 -m state --state ESTABLISHED,RELATED -j ACCEPT 
  $IPT -A FORWARD -s 10.8.0.0/24 -o eth0 -j ACCEPT 

  # Masquerade the openvpn network 
  $IPT -t nat -A POSTROUTING -s 10.8.0.0/24 ! -d 10.8.0.0/24 -o eth0 -j MASQUERADE 

  # Allow services 
  # Allow vpn on the default udp port 1194 
  $IPT -A SERVICES -p udp --dport 1194 -j ACCEPT 
  # Allow sshd on the default tcp port 22 
  $IPT -A SERVICES -p tcp --dport 22 -j ACCEPT 

} 

firewall_stop() { 

  # Polcies (permissive) 
  $IPT -P INPUT ACCEPT 

  # Flush the iptables 
  $IPT -F 

  # Delete the services custom chain 
  $IPT -X SERVICES 
} 

case "$1" in 
'start') 
  firewall_start 
  ;; 
'stop') 
  firewall_stop 
  ;; 
'restart') 
  firewall_stop 
  firewall_start 
  ;; 
*) 
  echo "Usage $0 start|stop|restart" 
esac

Give the firewall rc script executable permission:

# chmod +x /etc/rc.d/rc.firewall

and start it:

# /etc/rc.d/rc.firewall start

Now you will you need to do is to restart the OpenVPN service on the Server and restart the Client connection.

In the previous chapter we referred to a firewall you may include to protect your Openvpn Server. However this chapter refers to firewalls on the Client LAN that may block the VPN connection by blocking traffic on UDP port 1194.

In order to penetrate through the Client firewall your may want to try changing the port to 443 - normally reserved for https. Using TCP instead of UDP will also help. To make these change you will need to amend /etc/openvpn/server.conf of the Server, from

port 1194
proto udp

to:

port 443
proto tcp

and /etc/openvpn/client.conf of the Client, from

proto udp

remote servervpn.no-ip.org 1194

to:

proto tcp

remote servervpn.no-ip.org 443

The Server's firewall script would also need to be modified. Change these lines:

# allow vpn on the default udp port 1194 
$IPT -A SERVICES -p udp --dport 1194 -j ACCEPT 

to:

# allow vpn on the custom tcp port 443 
$IPT -A SERVICES -p tcp --dport 443 -j ACCEPT 

You also have to modify your Router's port forwarding to TCP port 443.

(1) http://en.wikipedia.org/wiki/OpenVPN

(2) https://wiki.archlinux.org/index.php/OpenVPN

(3) http://slackwiki.com/OpenVPN_smcr_2012

(4) http://www.no-ip.com

• Originally written by Chris Abela