QUESTIONS
1. What are the benefits of LAN segmentation using routers as compared to using
bridges or switches.
Using routers reduces the impact of broadcast/multicast because broadcast and
multicast frames are not forwarded by a router (by default). We can then create
multiple collision domains that are also separate broadcast domains. Remember
though that each broadcast domain must be part of the same Layer 3 address grouping
i.e. the same IP subnet (if we talk of IP), and multiple broadcast domain means we
need to have multiple subnets instead of only one, which becomes a problem if we are
short of addresses
Bridges and switches work at layer two and can quickly forward a frame to the
destination port, especially if cut-through technique is used where the switch doesn’t
wait until it receives the entire frame to start forwarding it. Routers on the other hand,
work at layer 3 and an entire packet must be received before routing decision can be
mad, which increases delay. The router also usually applies many logic steps before a
packet is routed, which can add latency.
Routers provide better manageability and functionality
Routers provides for multiple paths (routes) to destinations as opposed to bridges
which only specify on possiple output port.
2. What is meant by the terms classfull addressing and classless addressing and what
is the advantage of the latter as opposed to former?
Classfull addressing refers to the traditional IP addressing scheme where every
address belongs to a particular class e.g. class A, B or C.
With classless addressing instead, the address is allocated with a prefix mask
(the subnet mask) to identify the network portion of the address without regard for the
bit boundary at class A, B or C.
3. With reference to classless addressing what is meant by the term CIDR and what is
the advantage over traditional routing.
CIDR refers to routing protocols that transmit subnet masks in the routing updates.
4. Derive and document a simple procedure for the installation and configuration of
a new device?
5. Which features of the device did you evaluate practically, and why?
6. Distinguish between static and dynamic routing
Static routing is manually configured by the network administrator and it is
recommended for small networks with a fairly stable topology of for interconnecting
remote site via dial-on-demand links which are charged for the connection time. Static
routing in fact doesn’t involve the exchange of any information between routers.
Dynamic routing on the other hand, applies to more complex scenarios to
allow for scalability and easier manageability of the network.
7. List two reasons why you would you use static routing in this scenario. Do you
know any other reasons?
We don’t want to activate the ISDN line too often, which would be the case if we
were to use dynamic routing. Remember that dynamic routing protocols such as RIP,
IGRP etc regularly exchange information to keep track on changes in network
topology even if nothing has changed. This can result in useless usage of the available
bandwidth. Besides, in our scenario we are only connecting two sites and static
routing well applies here. Another reason for not using dynamic routing is also
usually security. If routers exchange information this can be tapped.
8. What is NAT and discuss its relevance to the scenario you are dealing with.
NAT stands for Network Address Translation and it’s a mechanism used for
translating one IP address into a different one. It can be useful in the following cases:
- To connect organizations that use address space issued to other organizations
on the Internet.
- To connect organizations that have used private addressing space, which
routers do not forward by default, and want to connect to internet
- To connect two organizations that have used the same private addresses (IP
addresses must be always unique).
- The organization wants to hide its address and is utilizing NAT as part of the
firewall capabilities
9. How many serial line protocols do you know of and which one is the OSI
recommended one?
HDLC (High-level Data Link Control) , SDLC (Synchronous Data Link Control),
PPP (Point –to-Point Protocol), SLIP (obsolete), LAPB (Link Access Procedure
Balanced) used mainly with X.25, LAPD (Link Access Procedure on the D channel)
used on D channels from router to ISDN switch for signalling, LAPF (Link Access
Procedure to Frame mode bearer services) used for Frame Relay signalling.
HDLC is the OSI recommended one and most of the other serial protocols are
based on it. It was originally devised for synchronous lines only but the ISO 3309
standard extends it to asynchronous lines too.
10.What are PPP and HDLC?
They are serial line protocols
11.What are the advantages of PPP over HDLC?
HDLC does not provide a mechanism to identify the type of packet it is carrying. For
example there is now way to distinguish IP traffic from IPX traffic (CISCO actually
solves the problem by introducing a proprietary protocol field)
PPP addresses HDLC limitation by using a 2-byte protocol field to
differentiate higher-level protocol traffic (for example x0021 is IP and x002b is IPX).
12.What are PAP and CHAP and what is the difference between them?
PAP (Password Authentication Protocol) and CHAP (Challenge Handshake
Authentication Protocol) are authentication protocols used on serial lines. PAP is
based on the validation of a username and password that are sent unencrypted across
the line; it doesn’t provide protection against playback attack. CHAP on the other
hand, is more secure because it uses a variable challenge value rather than a password
to authenticate the other party; this challenge value is unique and unpredictable and
provides protection against playback attack. The authentication process can be
repeated any time after the link is established. The latter also enhances security.