Chapter 3 - Fundamentals of WANs and IP Routing

Question 1

Define Leased Line

Answer 1

• A leased line is a dedicated private connection between an ISP and an organisation.
• It connects from the telco’s POP (Point of Presence) through multiple different devices in order to reach the organisation that is serves.
• It is a layer 1 service but the protocols that are used to transmit data over it are layer 2.

Question 2

What are two examples of data link protocols that are used to transmit data over leased lines?

Answer 2

HDLC (High-Level Data Link Control)
PPP (Point-to-Point Protocol)

Question 3

What is HDLC?

Answer 3

• HDLC (High Level Data-Link Control) - Layer 2 encapsulation protocol like Ethernet
• Unlike Ethernet it is more often used over a point-to-point topology.
• Transmission via HDLC doesn’t always need to provide a destination address in a point-to-point configuration as there is only one destination data can be sent to.

Question 4

How does an Ethernet WAN work?

Answer 4

An Ethernet WAN works similarly to how a leased line connects to a telco. The customer connects an Ethernet link to their own router with the other end connecting to a service provider’s PoP which contain’s that service provider’s Ethernet switch(es). The service provider can then use whichever technology they want to create the WAN service for the customer (e.g. EoMPLS).

Question 5

How does a router’s internal logic work for routing individual packets?

Answer 5

• A frame containing a layer 3 packet is received by a router
• The router uses FCS in the trailer of the frame to ensure that the frame has no errors
• Provided the frame has no errors, it is deencapsulated so that only the packet is left
• The router then compares the packet’s destination IP to its routing table and finds the route that best matches the destination IP. This route identifies the outgoing interface and potentially the next-hop IP
• The router then encapsulates the packet in a new frame appropriate for the outgoing interface and forwards the frame

Question 6

What does the Cisco proprietary HDLC header contain that the regular HDLC header does not?

Answer 6

A Type field.

Question 7

Define ARP and various ARP messages

Answer 7

• Address Resolution Protocol
• Dynamically learns the MAC address of an IP host connected to a LAN.
• ARP Request - A message sent to a host or router that states “If this is your IP address, please confirm your MAC address in response”
• ARP Reply - Lists the IP address from the ARP Request and the corresponding MAC address

Question 8

Define subnet

Answer 8

• A smaller partition of a larger network
• Two IP addresses separated from each other by at least one router must be in different subnets. If they are not separated by a router then they must be in the same subnet.

Question 9

Describe an IP Header

Answer 9

• 20 byte long minimum 60 byte maximum
• Layer 3 header
• Contains the source and destination IP address of a packet amongst other details.
• The destination and source IP addresses of a packet don’t change during the whole length of the transmission.

Question 10

How do many routing protocols learn IP routes?

Answer 10

• R1 adds a route to its routing table for each subnet that is directly connected to it independent of its routing protocol
• R1s routing protocol sends a routing update to R2 telling it about the routes in it’s routing table including the directly connected routes and routes learned from other routers
• R2 then adds the routes learned from R1 to it’s routing table and sends a similar update to R3
• After learning a new route from a R2, R3s routing protocol adds a route to its own routing table of how to get to the subnet connected to R1, where the next hop in this scenario would be R2

If multiple routes lead to the same subnet, the router will select the best route to take based on certain metrics.

Question 11

What is DNS and how does it work?

Answer 11

Domain Name System is a process used to resolve hostnames into matching IP addresses. One single DNS server doesn’t know every domain name and matching IP address so they all work together forwarding requests to each other until the query is answered correctly.

• PC1 attempts a connection to Server1 by typing in ‘Server1’
• A DNS query is sent from PC1 to its respective DNS server with the packet’s destination IP as the DNS server’s IP address
• The DNS server responds with a DNS reply to PC1 listing the IP address of Server1
• Now that PC1 knows Server1’s IP it can send a packet with the destination IP as Server1’s IP

Question 12

Define Ping

Answer 12

Ping (Packet Internet Groper) uses ICMP (Internet Control Message Protocol). PC1 will send an ICMP echo request to the IP address of PC2 and provided there are no networking issue, PC2 will reply with an ICMP echo reply.

Question 13

List and label the parts of an HDLC frame

Answer 13

Flag - Lists a recognizable bit pattern so that the receiving nodes realise that a new frame is arriving
Address - Identifies the destination device
Control - Defines the type of frame and its functionality (not necessary as much nowadays)
(Optional not always present) Information - Contains the data being transmitted by the frame
Type - Identifies the type of layer 3 packet encapsulated inside a frame
FCS - Uses error detection to check if the frame has experienced any transmission errors