• User account binding to the physical line, for example, using MAC-address of the network interface. Linux user has to configure network interfaces to establish such a connection;
• VPN-technologies allowing user authentication and secure connection. For example, PPTP, L2TP or IpSec;
• PPPoE (Point-to-Point Protocol over Ethernet) which this article is about.

PPPoE represents a network protocol for encapsulating Point-to-Point Protocol (PPP) frames inside Ethernet frames. PPPoE expands the capability of PPP allowing a virtual Point-to-Point connection over a multipoint Ethernet network architecture. Given protocol was published by IETF in 1999 and its specification is RFC 2516.

In fact PPPoE does not have important advantages in comparison to the above mentioned alternatives. However the following ones can be emphasized:

• It’s significantly more secure than MAC-binding and allows user to change network interfaces connected to the provider without appealing to provider’s customer support; still it slightly reduces the bandwidth of the network interface;
• Its configuration is simpler that the one of VPN-connection and requires less parameters to be introduced by user. In particular there is no need to configure IP-address or to specify the address of VPN-server.

Thus, most likely you will encounter PPPoE just because you ISP will provide the Internet via this technology. Although provider’s technical support generally can help you just with Windows, you can easily connect your PC with Linux operating system following our instructions.

The best way to use PPPoE on Linux is to do it with the help of pppd (Point-to-Point Protocol daemon) package, which supports PPPoE since version 2.4.2, and is usually pre-installed (in case it’s not, you can easily get and install it). PPPD is used to manage network connections between two nodes on Unix-like operating systems. The role of pppd is to manage PPP session establishment and session termination.

In addition to that there must be in-kernel PPPoE support. Most likely you already have it because it’s included in all modern distributions by default. However if you don’t, we will discuss how to add it a little bit later.

For now let’s see how you can set up a PPPoE connection using graphical interface (in this certain case we will use Ubuntu 9.04, though Ubuntu 9.10 beta is already available). Open: System – Preferences – Network Connection. Choose DSL tab and press “Add”:

Fill in the “Connection name” (anything you want), “Username” and “Password” fields, check or leave uncheck “Connect automatically”, “Show password” and “Available to all users” boxes:

NOTE: In case you are going to use this connection just for the Internet, you can leave “Service” field empty.

Choose Ipv4 Settings tab and make sure that the Method is Automatic (PPPoE):

That’s it! Now you can connect to the Internet without any problems.

However if you don’t really like using graphical interface or simply haven’t it (for example, on the server), there is another way to configure it.

First of all you need to check if there is in-kernel PPPoE support:

modprobe pppoe

In case the module wasn’t found, we need to create a kernel module. Make sure that you have kernel sources by going to the source directory – /usr/src/linux-x.x.x and run the following command:

make menuconfig

In the configuration window that will appear after the previous command open Network device support -> PPP over Ethernet (Experimental) and set “M” value there;

Or you can directly change .config (that can be found in the source directory - /usr/src/linux-x.x.x/):

CONFIG_PPPOE=m

Now we need to compile modules:

make modules

In case you get some compile errors like “field pppox have not defined in union…”, just as I did, try to make some changes in the following two files:

/usr/src/linux-x.x.x/drivers/net/pppoe.c

/usr/src/linux-x.x.x/drivers/net/pppox.c

- Add just after the comments (before any of #include directives) the following line:

#define CONFIG_PPPOE_MODULE

- Save the above mentioned files and retry compiling with

make modules

After successful compiling copy files

/usr/src/linux-x.x.x/drivers/net/pppoe.o

/usr/src/linux-x.x.x/drivers/net/pppox.o

(pay attention to “.o” extention!) to /lib/modules/x.x.x/kernel/drivers/net

Refresh the file with dependencies:

depmod -a

Now check one more time if the following command is working:

modprobe pppoe

In case it’s ok, we can go ahead with pppd configuration.

1) We need to add to /etc/ppp/chap-secrets our login and password.

If you plan to use pppd for more than one connection, than there is a necessity to introduce a unique ID along with login and password. For example, “PROV”:

ourlogin PROV ourpassword *

(of course, instead of “ourlogin” and “ourpassword” we need to write true values)

2) We have to create a file (let it be “myprov”) in /etc/ppp/peers.

Here are the things we need to write in this file:

user ourlogin (ourlogin – the same login as we introduced in chap-secrets)

plugin rp-pppoe.so

eth0 (your interface via which you want to receive pppoe)

noipdefault (in case your ISP provides dynamic IP)

usepeerdns (to receive DNS from provider)

defaultroute (while connecting to the Internet the provider’s router will be used as a gateway by default)

persist (fixed connection)

remotename PROV (PROV – an ID we introduced in chap-secrets)

3) Finally we need to save everything. To call the connection we use the command:

pppd call myprov

If pppd displays that it can’t load, for example, /usr/lib/pppd/2.4.3/rp-pppoe.so, we need to check where pppd is:

whereis pppd

-  After that we should make a link in /usr/lib/pppd/2.4.3 to our rp-pppoe.so:

ln –s [our rp-pppoe.so] [/usr/lib/pppd/2.4.3/rp-pppoe.so]

If pppd loads, but fails to create PPPoE socket, check log, maybe there are some messages like:

modprobe: can’t locate module net-pf-24

-  If there are, we need to write in /etc/modules.conf:

alias net-pf-24 pppoe

No doubt using PPPoE in Linux is as simple as it is in other operating systems. Graphical interface allows you to connect it in a very fast and intuitive way. However if you want be able to solve different kind of problems that may appear, it’s a great idea to make use of command line. Anyway it won’t take you more than 5-10 minutes to setup the connection, so check it out right now in order to see how easy it is.

Related posts:

1. Setting Up Network Interfaces in Linux
2. Setting Up SSH clients on Windows Mobile Devices
3. IPv6 advantages and its configuration in OS Linux

IPv6 has a vastly larger address space than IPv4. This results from the use of a 128-bit address, whereas IPv4 uses only 32 bits. The new address space thus supports 2^128 (about 3.4×10^38) addresses. Thus IPv6 eliminates the need for network address translation (NAT), which gained widespread deployment as an effort to solve the problem of IPv4 address exhaustion.

Of course, IPv4 address exhaustion is not the only problem that can be fixed by IPv6. Moreover IPv6 has other important improvements to IPv4 such as routing, network autoconfiguration, network security and etc.

In addition to the above mentioned improvements here the significant advantages of IPv6 which can be very important for the end users:

1. Static IP-addresses for your computers

Nowadays, in case you don’t have direct IPv6, the best way to connect to IPv6 is to register an account at a tunnel broker’s site. A tunnel broker is a company that offers service (for free) of transferring your traffic from IPv4 to IPv6 and backwards.

Using this method user gets his/her own IPv6 subnet, which is tied not to the current IPv4 address, but to his/her account (login and password). Thus you have the opportunity not just to get the range of IPv6 addresses, but to use them even if you change your IPv4 provider.

Moreover, every user gets a subnet /64, which is enough to connect to the network 2^64 devices and assign them static Internet addresses. This means that you can access any computer of your network directly (via its address) without the necessity to use NAT.

Some users may think that security problems will appear in this case, isn’t it? In fact there is no reason to worry. The only thing you should do is to configure your firewall, so it doesn’t pass connections from the Internet to the local network, except the ones that were allowed.

2. Higher speed for torrents downloading

BitTorrent Protocol is designed in such a way that users who do not have the opportunity to receive incoming connection (provider’s NAT is used) can “torrent” the files just from the users who have own IP address (NAT is not used). That’s why the growing shortage of IPv4 addresses will cause the decrease in amount of torrent peers and seeds.

If you use IPv6 there is no need to worry about this problem, because IPv6 users can get own IP addresses even in case their IPv4 provider uses NAT for them. Of course, you should check if your favorite tracker supports IPv6.

3. Long-lived connections that remain even after IPv4 address changing

   Internet access control systems of some providers work in such a way that maximum duration of an Internet connection is equal to 24 hours, after which a forced disconnection takes place. During the repeated connection new IP-address is given. This can happen while you are chatting, playing online games, use different online services and etc.
   If you connect using IPv6, there won’t be a disconnection while IPv4 address is changing, because IPv6 remains the same. The only thing that will happen is discontinuance of data transfer for 15 seconds – the time needed for a tunnel broker to get the info about new IPv4.

However despite all the improvements and advantages the worldwide penetration of IPv6 protocol is still less than one percent of Internet-enabled hosts in any country; though IPv6 has been implemented on all major operating systems in commercial, business, and home consumer environments.

For now let’s discuss the alternatives of IPv6 connection:

1. Direct connection

Pros:

- The most reliable one

- It’s very easy to configure

- The lowest ping and high speed

Cons:

- In case it’s not offered by your Internet provider there is no way to connect it

- At the moment the amount of providers who offer IPv6 is pretty limited

2. 6to4

6to4 addressing represents a type of IPv6 address created by your IPv4 address. It allows you to set up tunnels over the IPv4 Internet automatically. Thus all the boxes behind your firewall will have IPv6 addresses. In case you would like to learn more please visit this page.

Pros:

- There is no need to register any account

- Optimal (the closest) gateway is chosen automatically by anycast method

Cons:

- In case of dynamical IPv4, its change will cause the change of IPv6 as well

3. Teredo

Teredo is an address assignment and automatic tunneling technology that provides unicast IPv6 connectivity across the IPv4 Internet, when IPv6/IPv4 hosts are located behind one or multiple IPv4 network address translators (NATs). If you need more info about Teredo, you can get it here.

Pros:

- It’s very easy to setup (without the necessity to register an account)

- There is a direct connection with ping between two Teredo users that is no worse than direct IPv6 connection

- In Windows 7 it’s turned on by default

Cons:

- In case of dynamical IPv4, its change will cause the change of IPv6 as well

4. Tunnel Broker

Pros:

- It gives static IPv6 range, which does not depend on your Internet provider and does not change if IPv4-address is changed.

- Some brokers give you the opportunity to easily (via web-interface) delegate reverse DNS to your DNS-servers

Cons:

- You need to create an account on the broker’s site

- Some brokers have to be informed when your IPv4 is changed (still this can be automated)

In modern GNU/Linux distributions there is an opportunity to automatically get configured and working IPv6-tunnel by entering the only command (even if your provider does not support IPv6 or provides access to the Internet just via NAT). To do so you have to install a program called Miredo (http://www.remlab.net/miredo/). But before that make sure that the usage of UDP-port 3544 is not prohibited by your firewall.

The main characteristic feature of given method is that your IPv6-address will be changed each time you change your IPv4-address. In case you would like to get a static IPv6-address (it possible, moreover, absolutely free of charge), please use the tunnel broker method, which will be described a little bit later.

Now let’s go ahead with the IPv6 installation…

First of all we need to make sure that there is no access to IPv6 on the given computer:

# ping6 ipv6.google.com
connect: Network is unreachable

Now we need to install Miredo:

sudo apt-get install miredo
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following NEW packages will be installed:
miredo
0 upgraded, 1 newly installed, 0 to remove and 0 not upgraded.
Need to get 167kB of archives.
After this operation, 430kB of additional disk space will be used.
Get:1 http://localhost lenny/main miredo 1.1.5-1 [167kB]
Fetched 167kB in 0s (1148kB/s)
Selecting previously deselected package miredo.
(Reading database ... 34093 files and directories currently installed.)
Unpacking miredo (from .../miredo_1.1.5-1_amd64.deb) ...
Processing triggers for man-db ...
Setting up miredo (1.1.5-1) ...
Starting Teredo IPv6 tunneling daemon: miredo.

Check IPv6 again:

# ping6 ipv6.google.com
PING ipv6.google.com(fx-in-x68.google.com) 56 data bytes
64 bytes from fx-in-x68.google.com: icmp_seq=1 ttl=58 time=135 ms
64 bytes from fx-in-x68.google.com: icmp_seq=2 ttl=58 time=135 ms
64 bytes from fx-in-x68.google.com: icmp_seq=3 ttl=58 time=135 ms
64 bytes from fx-in-x68.google.com: icmp_seq=4 ttl=58 time=135 ms
^C
--- ipv6.google.com ping statistics ---
4 packets transmitted, 4 received, 0% packet loss, time 3006ms
rtt min/avg/max/mdev = 135.393/135.547/135.631/0.091 ms

It’s working! Now you can use any of available IPv6 services and enjoy the fact that your computer belongs to the future Internet.

Related posts:

1. Setting Up PPPoE Connection in Linux
2. Gibraltar v3.0
3. Setting Up Network Interfaces in Linux