WAN – Point-to-Point Protocol (PPP)

In this article I will talk about the Point-to-Point Protocol(PPP) used in point-to-point communications. PPP is one of the most used WAN technologies in data networks all around the world. This type of serial connection is mostly used to connect LANs between each other or to connect an enterprise network with a Service Provider. A point-to-point connection between a company and an ISP (Internet Service Provider) is also known as a leased line. PPP offers support for many WAN technologies like Frame Relay or ATM, but also provides a multi-protocol architecture for TCP/IP, Appletalk or IPX. I will show you how to configure Point-to-Point connections, how to troubleshoot them and also how to configure PPP and CHAP authentication modes.

   I’ve told you that point-to-point connections use serial communications, but what exactly are those and what is the difference between a serial and a parallel communication? In serial communications, bits are sent one after the other while in parallel communications, multiple bits can be sent together using different lines. You would probably say that parallel communications are preferred because they offer increased speed since bits can be sent faster than in serial communications. Parallel communications are susceptible to clock skew and interference. Clock skew simply means that bits that are sent together from one point do not arrive at the same time at the other end. This happens because both ends of communication must synchronize when transmitting information over the medium. You can clock serial communications to achieve faster speeds than parallel communications. In parallel communications, bits can interfere between them and many can get dropped because of this. Simply put, serial communications are preferred in point-to-point links because they require less physical resources (wires and cables), can provide faster speeds than parallel communications, support longer cable distances and also they can be better isolated so that data transfers do not suffer from interference. There are many point-to-point WAN standards used today but among the well known standards are (I will not talk about each of them but I’ll put some interesting links if someone is interested) :

HSSI (High-Speed Serial Interface (HSSI) – HSSI – http://en.wikipedia.org/wiki/High-Speed_Serial_Interface

   The main concept used in Point-to-Point connections is the TDM or Time Division Multiplexing. In this layer 1 concept, every node that wants to transfer data over the medium receives a timeslot in which it can transfer bits over the physical connection. A multiplexer is responsible for allocating timeslots to the users and this devices also reassembles each data stream. Remember that these timeslots are interleaved in the physical channel (we’ve talked about the interleaving process in an earlier article from the networking fundamentals section). In the first TDM implementations, timeslots were 8 bits long but this concept had a problem. When a user didn’t had nothing to send over the channel, the TDM mechanism would still allocate a timeslot for that particular user. To address this issue, statistical time-division multiplexing was invented. In this technology, a buffer was created in which data can be stored when there is high traffic onto the medium. By using this method, STDM ensures that the physical channel doesn’t remain idle when there is no information transmitted.
   You know from earlier articles that point-to-point connections use two devices, a DTE and a DCE. The DTE is part of the CPE (customer premises equipment) while the DCE is located in the ISP’s network. The DCE, which can be a modem or a CSU/DSU device, provides the clock signal for the serial communication. Unfortunately I don’t have pictures with WAN connectors to show you and I cannot use pictures that are taken from some other places, but if you are interested you can look for some of the most used serial connectors like DB-60, Smart Serial, V.35, X.21, EIA-530 etc. I don’t think is important to know these connectors or their role for the CCNA exam, but you can make a general idea by searching on www.google.com for them.

   There are many WAN encapsulation protocols used in serial connections. We will study some of them later, but for now I just want to point out the most used encapsulation protocols today:
HDLC (High-Level Data Link Control) – default encapsulation protocol used in point-to-point communications (this protocol is enabled by default on all Cisco devices). It is a protocol that provides connection-oriented and connectionless services. This protocol uses ACK messages when sending and receiving frames and uses synchronous serial communications. HDLC adds a special flag that signalizes the beginning and the end of a frame, the flag is 8 bit long and it’s 01111110. When there are five consecutive 1s in a stream of bits, HDLC will insert another 1 bit in order to make sure that the flag is inserted in the right spot. It is pretty simple to change the encapsulation protocol used in serial connections. This is done by typing encapsulation hdlc from the interface configuration mode:

interface encapsulation

Remember that HDLC is the default encapsulation mode used by Cisco routers. To verify the encapsulation protocol, type show interfaces serial [number]:

show interface command
PPP –  “is a data link protocol commonly used in establishing a direct connection between two networking nodes” from Wikipedia http://en.wikipedia.org/wiki/Point-to-point_protocol. We will talk about the PPP protocol later in this article.
Frame Relay – data link protocol that uses Virtual Circuits (VC) to sent/receive data. It is an improved version of the X.25 protocol, I will talk about Frame Relay in a future article.
SLIP – point-to-point protocol that uses TCP/IP for transmitting data.
The point-to-point protocol is used when you want to connect non Cisco devices between each other. This is a serial protocol known by all networking devices and it has some features that cannot be found in HDLC. It has a feature to detect the link quality and also it supports authentication using the PAP or CHAP protocols, we will talk about these two authentication protocols later in this article. PPP uses the HDLC protocol to encapsulate IP datagrams in point-to-point connections. PPP has another protocol called the LCP (Link Control Protocol) protocol used for configuring, establishing connections and for checking the state of point-to-point links. Another component of the PPP protocol is the NCP or the Network Control Protocol. NCPs are used to configure network protocols like IP, IPX or Appletalk, over the serial communication.
PPP uses the last three OSI layers, the physical, data-link and the network layer. At the physical layer, PPP can be configured in many serial interfaces like synchronous, asychronous or HSSI. The LCPs are used to establish, terminate, configure and test connections. You’ll have to know for the CCNA exam that the LCP layer from the PPP protocol is used to set the error detection, compression and authentication mechanisms. The NCP layer is used by PPP to encapsulate different network protocols. When a PPP connection is made, three phases must be done: connection, establishment, link quality and the network protocol determination. There is much to talk about the NCP or LCP operation, check the following link from tcpipguide for further details http://www.tcpipguide.com/free/t_PPPLinkControlProtocolLCP.htm.
The Point-to-Point protocol offers the following options:
authentication – can provide two authentication mechanisms, PPP and CHAP.
error detection – by using magic numbers and quality numbers, PPP ensures that the link doesn’t contain errors.
multilink support – it is a mechanism used to load balance traffic over multiple physical PPP links.
compression – using the Stacker and Predictor protocols, PPP can reduce the size of frames.
PPP callback – a security mechanism in which one side must call the other side and by answering, the PPP link is established.
To configure PPP on a Cisco device, first set the encapsulation type to PPP from the interface configuration mode:

interface encapsulation
Now, if we want to set the quality of the link, we just have to type in ppp quality [percentage]. The quality is calculated by the number of packets send and received. If the link quality doesn’t meet our expectations, then PPP will shutdown the link. To set the multilink option, simply type in ppp multilink from the interface configuration mode. The compression option is enabled by the compress [predictor/stac] command. To verify your PPP configuration use the show interfaces, show running-config and the show interfaces serial commands. To troubleshoot PPP you can activate the debug ppp feature.
   I’ve written earlier that PPP supports two authentication mechanisms, PAP (password authentication protocol) and CHAP (Challenge-Handshake Authentication Protocol):
PAP is a simple implementation of an authentication mechanism, the two devices participating in the PPP link establishment must first authenticate each other using a username and a password. This is a two-way method of authentication. The first router must send it’s credentials to the second router which will grant or deny the connection. PAP will sent the credentials in plain text, this is why it’s not a secure method of authentication because it’s susceptible to interception. After the link is established, PAP will not ask again for the credentials. Read more about the PAP protocol, in this article from Wikipedia: http://en.wikipedia.org/wiki/Password_authentication_protocol.
CHAP on the other hand, uses an encryption algorithm (MD5) when sending credentials and it will also ask for the username and password periodically. CHAP is a three-way method in which the second router must first send a challenge message and then the first router must respond with a hash value of the encrypted credentials. In the third step of the CHAP authentication process, the router will check the received credentials and it will deny or accept the connection. You can configure the users/passwords locally on the routers or use a AAA/TACACS (a server which is used to authenticate users).
   We will configure a PPP connection between two Cisco routers, R1 and R2, using the PAP authentication method. We will first configure PAP on router R1:
we will first need to configure a username on R1: R1(config)#username R2 password test
The next image displays all the configuration commands needed to configure PAP on this router:
PPP configuration commands
The same commands must be entered on router R2: R2(config)#username R1 password test
After the username has been configured, the PAP protocol must be configured on this router too:
PPP authentication
To configure CHAP on these two routers, simply create the usernames/passwords and then type in ppp authentication CHAP from the interface configuration mode. If for some reason the ppp authentication fails, you can enable the Debug ppp authentication mode.
   I hope I’ve covered all the main components that make up the PPP protocol. If you think there is something more to add here, don’t hesitate to leave a comment or post any question that you have. I wish you all the best, please share this article to others and stay tuned for more to come.

2 thoughts on “WAN – Point-to-Point Protocol (PPP)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s