IP Networking

1
IP Networking
2
What is a Network?
A connection between two or more computers so
they may exchange information/resources. Using a
network medium, computers pass information from
one to another in an organized fashion using
communication rules governed by protocols.
3
Network Medium
Refers to the tool used to carry the transmitted
signals from one computer to the next. Typically
Unshielded Twisted Pair cabling is the most
common (aka 10BaseT or 100BaseT).
4
Protocol

• The language, or set of rules, that govern the
  communication between two devices.
• The protocol determines the following
• the type of error checking to be used
• data compression method, if any
• how the sending device will indicate that it has
  finished sending a message
• how the receiving device will indicate that it
  has received a message

TCP/IP, IPX/SPX, Apple Talk, NetBEUI, etc.
5
Network Types
1. LAN The most common type of network is the
Local Area Network in which computers situated in
a common physical enclosure are connected
together. A typical office scenario would be the
most common example.
6
Network Types
2. WAN When two buildings need to be connected
together a Wide Area Network is needed.
Typically joining two LANs together creates a
WAN.
How do we join them?
7
Traditional Methods
T1 T3 Fiber DSL
Modem
8
WaveRider Methods
Wireless
Wireless
9
Our NCL family of products is used to join two
LANs and create a WAN
10
while our LMS product line provides a wireless
Last Mile Solution for ISPs and home/small
business users.
11
WaveRider Product
Two Network Interfaces
RF Interface (SMA connector)
Ethernet Interface (10BaseT)
12
RF Perspective
Issues such as antenna selection, cabling,
distance, Line of Sight, configuration, etc. must
all be considered. These topics covered in the RF
Networking section.
Ethernet Perspective
Issues such as what type of protocol to use,
medium and configuration must be considered.
13
How to Implement?

• Two options
• Bridge Mode
• Route Mode

• How do I choose??

14
Bridge Mode
Used when connecting two segments of the same
network together via the wireless link. In Bridge
Mode, the NCL product supports all protocols.
TCP/IP NetBEUI IPX/SPX AppleTalk
15
Route Mode
Used when connecting two different networks
together. The NCL/LMS products support TCP/IP
when functioning in Route Mode. Take advantage of
Route Tables to route packets from one network to
another.
TCP/IP
Network ID 192.168.10.0
Network ID 10.32.0.0
16
TCP/IP The Internet Protocol
Due to the popularity of the Internet and
internet services such as e-mail and web pages,
the TCP/IP protocol has become the protocol of
choice for most networks. It also allows the
network administrator to use SNMP, DHCP and other
services to help manage the network.
TCP/IP
Transmission Control Protocol/Internet Protocol
17
What is TCP/IP?
Suite of protocols that, when used together,
allow computers to pass data between one another.
Each protocol in the suite performs a specific
function in the communication process.
- used for connection based communication
TCP
- used for resolving IP addresses to MAC
addresses
ARP
- used for remotely monitoring network devices
SNMP
- used for addressing and routing data to the
destination
IP
DHCP, RIP, ICMP, UDP, FTP, HTTP, etc.
18
In 1978 the International Standards Organization
introduced an architectural model that would
allow all networks to be able to communicate with
each other. This model is called the Open
Standards Interconnect Reference Model
(OSI). All manufacturers have developed their
product according to this standard.
Understanding the OSI model will help you to
understand how communication on a network happens.
19
The OSI Reference Model
Represents a layered approach to networking.
Each layer of the model handles a different
portion of the communications process. The OSI
Model simplifies how hardware and software work
together, as well as providing a specific method
of how components should function. The OSI model
consists of 7 Layers.
20
(No Transcript)
21
Physical Layer (Layer 1) Converts incoming
signals into bits and converts outgoing bits into
signals by managing the interface between the
computer and the network medium. Data Link (Layer
2) Responsible for creating the frame type used
on the network (ie. Ethernet vs. Token Ring).
Also provides error checking to determine when
data needs to be re-transmitted. Network Layer
(Layer 3) Defines protocols for routing data by
opening and maintaining a path on the network
between systems to ensure that data arrives at
the correct destination node.
22
Transport Layer (Layer 4) Controls the movement
of data between systems, defines protocols for
structuring messages, and supervises the validity
of transmissions by performing error
checking. Session Layer (Layer 5) Coordinates
communication between systems, maintaining
sessions for as long as needed and performing
security, logging, and administrative
functions. Presentation Layer (Layer 6) Defines
how information is formatted for display or
printing, how data is encrypted, and translation
of other character sets.
23
Application Layer (Layer 7) The highest layer of
the model, defines the way applications interact
with the network. As data is sent through the
model, a header is added at each layer. On the
receiving end the data is passed from Layer 1 to
Layer 7 with the headers being stripped off as
they go until the Application receives the data
to be processed.
24
TCP/IP Model
TCP/IP has its own reference model which relates
to the OSI Model. By understanding the two
models it is easy to see how the various
protocols participate in the communication
process.
FTP, SNMP, HTTP, SMTP, Telnet, PING
TCP UDP
ICMP IP ARP
25
Network Addressing
For a packet of information to reach its
destination it must be addressed with the
destinations address, much the same way a
regular mail letter needs an address before you
put it into the post box. Where peoples homes
have street addresses, with IP networking each
computer has an IP address.
10.32.0.24
255 Consumers Road
26
On any given network (or street) no two computers
(or homes) can have the same address. If so,
then where would the packet (or mail) be
delivered!?!?
260 Consumers Road
255 Consumers Road
265 Consumers Road
10.32.0.10
10.32.0.100
10.32.0.24
27
An IP address consists of two parts a NetID
portion and a HostID Portion A house address
consists of two parts a street name and a house
number
255 Consumers Road
10.32.0.24
28
Obviously there could be two similar addresses in
the world, the only difference being the cities
each is located. Identifying the city tells the
mail carrier how to handle the packet. In the
computer world the same holds true. The Subnet
Mask associated with the computer defines which
network it belongs to.
255 Consumers Road Toronto, Canada
265 Consumers Road Melbourne, AUS
29
IP 10.32.0.100 SM 255.255.255.224
IP 10.32.0.24 SM 255.255.255.224
Network 10.32.0.0
Network 10.32.0.96
The Subnet Mask determines what NetID the
computer has.
30
IP Rulesso far

• All computers on a given network must have same
  NetID
• No two computers can have the same HostID
• All computers on the same network must have the
  same subnet mask

31
IP Address
Every device on a TCP/IP network requires a
unique IP address. The IP address is commonly
represented in Dotted Decimal Notation.
207.23.175.2
In reality, the IP address is really a 32-bit
binary number. A computer would view the above
IP address as
11001111.00010111.10101111.00000010
Each quarter of the IP address is called an
Octet, because each quarter is made up of an
8-bit binary number.
32
Binary Numbers
Each bit in a binary number is assigned a decimal
value. The more bits available in a binary
number, the larger the decimal value of the
number becomes.
128 64 32 16 8 4 2 1

207
1 1 0 0 1 1 1 1
Tip When converting from decimal to binary,
always start from the left and add bits to the
right.
33
Review Test 1
34
Movie

• Warriors of the Net

35
Application Layer Protocols
36
HTTP FTP Telnet SNMP Remember the
Application Layer of the TCP model is responsible
for interacting with applications. Many
applications have a certain protocol they rely on
to get the job done. Some common applications
and their associated protocols would be Web
Browser HTTP SNMP Manager SNMP E-Mail Program
SMTP
37
Hyper Text Transfer Protocol (HTTP) is the
protocol used to transfer the data used to create
Web Pages in your browser. To the end user the
use of the protocol goes unnoticed, but to the
application it is very important. Another common
protocol is the File Transfer Protocol (FTP).
When transferring files over the Internet this
protocol is used to create a connection oriented
session to transfer the data. TCP is used to
create the session and guarantee delivery of
packets.
38
Telnet is a protocol commonly used to remotely
administer Routers and other network devices from
a remote location. It allows you to establish a
session with the other machine and emulate
keystrokes on the machine from your remote
computer. All components of the LMS are
manageable through Telnet sessions. The LMS and
NCL products are both configurable through a
Telnet session to the unit.
39
Simple Network Management Protocol (SNMP) is used
to remotely monitor or modify devices on your IP
network. Unlike Telnet where you are actually
emulating keystrokes on the unit, with SNMP you
are polling devices for information. Depending
on the security you set the units up with you
can 1. View information on the unit 2. View
and edit information on the unit
SNMP relies on a Management Station to ask
questions to Agents which respond to the
questions appropriately.
40
The data that a Management station requests from
an agent is held in a MIB (Management Information
Base). Basically it is a list of questions that
the management system can ask an agent. Managers
and Agents use the MIB tables as a lookup to send
data back and forth. Both the LMS and the NCL
both have their own MIBs which are installed at
the management station. That way the management
station will know what information it can poll
the agent for.
41
Transport Layer Protocols
42

• TCP UDP
• Remember the Transport Layer of the TCP model is
  responsible for delivery and movement of packets.
• There are two ways in which a packet can be
  delivered to another computer.
• Guaranteed with some sort of acknowledgement of
  delivery (ie. Registered Mail)
• Non-Guaranteed with no acknowledgment of delivery
  (ie. Regular Mail)

43
When the delivery of packets needs to be
guaranteed the Transmission Control Protocol
(TCP) is used at this layer to establish a
connection oriented session between the two
computers. The session tracks packets by
monitoring when a packet is sent, in what order
it is sent, and by notifying the sender when it
is received. If for some reason packets are
lost, the sending machine will resend the data.
All of this chatter between the two computers
causes TCP to have more overhead than its
counterpart UDP.
44
The User Datagram Protocol (UDP) is a
non-connection based protocol that does not
require a session to be established before
transmitting data. Packets are not guaranteed
and are also not retransmitted if lost in
transit. Because of the low overhead of UDP it is
most commonly used with applications that depend
on speed and where the guarantee of delivery is
not very important. SNMP applications and
real-time audio/video applications are good
examples of applications that use UDP as a
carrier.
45
TCP is typically used for transferring large
amounts of data, where the application requires
acknowledgement that the data has been received.
SNMP applications and real-time audio/video
applications are good examples of applications
that use UDP as a carrier.
46
Internet Layer Protocols
47
IP ARP DHCP ICMP Remember the Internet
Layer of the TCP model is responsible for
addressing and routing of packets.
Every packet a computer generates must be
addressed with an IP address. There are two
possible types of destination IP addresses a
packet could possibly have

• Local
• Remote

48
Local When a computer is sending data to a
computer on the same logical network as it is
located on, the two computers are said to be
local to each other.
Remote When a computer is sending data to a
computer on a different logical network than the
one it is located on, the two computers are said
to be remote to each other. In this case, a
router must be present on each network to allow
data packets to travel back and forth between the
two.
49
NetID
The NetID is what the sending computer looks at
to determine if the destination is local or
remote to it.
IF (Destination NetID) (Local NetID) then
destination is local
IF (Destination NetID) lt gt (Local NetID) then
destination is remote
How do we determine the NetID?
50
ANDing
Upon boot up, all computers determine their own
NetID by ANDing their IP address and their Subnet
Mask.
11001000.01100100.00001010.10001100 11111111.11111
111.11111111.11100000 11001000.01100100.00001010.1
0000000
IP 200.100.10.140 SM 255.255.255.224
1 1 1 1 0 0 0 0 0
200.100.10.128 a Network ID
51
The sending computer also ANDs the destination IP
address of each packet with its own Subnet Mask
to determine whether the packet is local or
remote.
11001000.01100100.00001010.10000101 11111111.11111
111.11111111.11100000 11001000.01100100.00001010.1
0000000
Dest IP 200.100.10.133 SM 255.255.255.224
200.100.10.128 a Local Network ID
Dest IP 200.100.10.165 SM 255.255.255.224
11001000.01100100.00001010.10100101 11111111.11111
111.11111111.11100000 11001000.01100100.00001010.1
0100000
200.100.10.160 a Remote Network ID
52
Why did we do that?
By determining if the destination was Local or
Remote, we are then able to address the packet
properly. When local, we stamp the packet with
the destination IP address. If remote we not only
stamp it with the destination IP address, but we
also stamp it with the default gateway (aka
Router) IP address.
Now the Internet Layer can pass the packet to the
lower layers of the OSI Model right??
53
One More Step
Unfortunately at the next layer (Data Link), a
different addressing scheme is used. This layer
uses the MAC address to determine if packets are
to be forwarded to the upper layers, therefore we
must also address each packet with the MAC
address of the final destination.
Now that we know the destination IP address, how
do we determine the MAC address needed?
54
ARP Address Resolution Protocol
Address Resolution Protocol allows a computer to
send a broadcast on the network which basically
says I need to send a message to 192.168.9.2.
Can the computer with this IP address please
respond to this message and send your MAC address
to 44-32-F9-00-00-01
55
This message is sent as a broadcast on the
network so that every computer on the network
receives it. A broadcast message is sent with
the MAC address of FF-FF-FF-FF-FF-FF. Every
computer reads the message, but only the computer
with a matching IP address will respond. After
the resolution takes place and a response is
received, the requesting computer will then add
the resolved MAC address to its ARP table, for
future data transmission.
56
A typical ARP table contains the IP and
corresponding MAC address. By storing this in
memory, if the MAC has already been resolved, a
broadcast message (which slows down the network)
does not need to be sent. This table is usually
flushed every few minutes, but can be made to
stay resident in memory.
57
If IP has determined the destination to be Local,
ARP will look for the MAC address of the
destination. If IP has determined the destination
to be Remote, ARP will look for the MAC address
of the Default Gateway. Given the previous
information, we can now conclude how a computer
communicates with both Local and Remote hosts.
58
Resolving a Local IP Address
1. Determine if IP is Local or Remote
2. Is there already an entry in ARP table? If
not, send an ARP broadcast.

• Once address resolved, address the data packet
  with MAC and send.

59
When the IP is Remote
1. Determine if IP is Local or Remote

• If remote, the source will send the data packet
  to its Default Gateway (Router), therefore the
  MAC of the router needs to be resolved.

• An ARP is sent requesting the MAC for the Router.
  (All nodes on a network would be preprogrammed
  with the IP of the router)

• Once the MAC of the router is resolved, the data
  is sent to the router. The data packet contains
  the IP of final destination. The router then
  goes through the same process of determining
  whether that IP is local or remote to it.

5. This process continues until the final
destination is reached.
60
In all networking environments one rule must be
followed..
No two computers can have the same IP address at
any one time
An IP conflict occurs in this situation, with
neither computer being able to communicate on the
network. There are ways of assigning IPs without
causing conflicts.
61
The best way to resolve any potential IP
conflicts on your network is with the use of a
DHCP Server.
Dynamic Host Configuration Protocol
A DHCP Server is a computer on your network that
hands out IP addresses to hosts as they come
online on the network. TCP/IP is a high
maintenance protocol, and the DHCP server takes
much of the management away from the Network
Administrator.
62
IP Configuration Options
63
Using a DHCP Server will allow automatic
configuration of all these parameters (and more)
without any manual intervention.
64
A client broadcasts a request on the network
which only a DHCP server will reply to. The
request basically says The computer with MAC
address ?? Needs an IP address. Please send me
one!
The DHCP server responds with at least an IP
address and Subnet Mask. Other options are
configurable as well at the Network
Administrators option.
65
Broadcasts
An important thing to remember is that, by
default, most routers will not forward broadcast
messages. If your DHCP server is located on the
other side of a router, generally a DHCP Relay
Agent must be installed on the network to act as
an intermediary between the client and the
server. The Relay Agent is already configured
with an IP address, and hence can get directly to
the DHCP Server by knowing the IP Address of the
server.
66
WaveRider Products
Both the LMS and the NCL (when in route mode) can
be configured to act as DHCP Relay Agents,
thereby avoiding any issues with DHCP broadcast
messages across networks.
67
As most users know, sometimes things just dont
go right when logging onto a network, or surfing
the web, etc. Most of us have seen messages that
resemble the following
68

• To report any error messages regarding IP and
  the delivery of packages, the ICMP (Internet
  Control Message Protocol) is used. There are two
  basic types of messages error and query
• Error messages might include
• Destination Unreachable
• Redirect
• Time Exceeded

69

• Query messages might include
• Echo Request
• Echo Reply
• PING (Personal Internet Groper) is a protocol
  that uses ICMP to verify that a particular IP
  address exists on a network. One of the most
  common and easy to use troubleshooting methods
  available.

70
IP Addresses
71
How does a company pick its IP address?

• Two options
• Public IP addressing
• Private IP addressing

Most companies are taking advantage of the
advanced features available through routers and
are using Private IP addressing. This reduces
both the setup and administrative costs of the
network.
72

• There are still rules that must be followed when
  choosing your own IP addresses, especially if you
  choose to connect to the internet eventually.
• RFC 1918 clearly defines which NetIDs are allowed
  in a private addressing scheme
• 10.0.0.0 - 10.255.255.255 /8
• 172.16.0.0 - 172.31.255.255 /16
• 192.168.0.0 - 192.168.255.255 /16

73
The disadvantage to this addressing scheme is
that it requires an organization to use a Network
Address Translator (NAT) at their router for
global Internet access. However, the use of the
private address space and NAT make it much easier
for clients to change their ISP without the need
to renumber their network.
74
Public IP addresses are assigned and administered
by the Network Providers (Telephone Companies,
ISPs) and the whole orchestration of addressing
is ultimately managed by a single organization
known as Network Solutions. For the most part,
IP Addresses are given in chunks to Internet
Service Providers who in turn hand them out to
their clients.
75
Determining the Address Class
By simply looking at the first octet of an IP
address, its class is easily determined. This
will help answer other questions such as how many
HostIDs are available, what is the default Subnet
Mask, etc. Use the following chart to read the
first octet
76
After determining the IP Class, we can now
determine how many octets (or bits) are allocated
to the NetID and HostID.
Remember Each octet contains 8 bits
77
With each IP Class there is also a default Subnet
Mask associated with each. They are as
follows Class A 255.0.0.0 Class B
255.255.0.0 Class C 255.255.255.0
a b.c.d a.b c.d a.b.c d
As can be seen, there is a direct relationship
between the default subnet mask and the number of
octets available for the NetID and the HostID.
78
By reading the subnet mask in binary
SM 255.255.255.0
11111111.11111111.11111111.00000000
and knowing that the first 3 octets are for the
NetID and the last octet is for the HostID with a
Class C Address we can deduce the following

• a 1 in the SM means the corresponding bit in
  the IP address is used for the Network ID
• a 0 in the SM means the corresponding bit in
  the IP address is used for the Host ID

79
Example
IP 207.23.175.2 11001111.00010111.10101111.
00000010 SM 255.255.255.0 11111111.11111111.111
11111.00000000
By following the rule, we can clearly identify
both the NetID and the HostID. NetID
207.23.175.0 HostID 2
80
How many Hosts??
Now that we know how to determine the class and
interpret the Subnet Mask, we can use that
information to determine how many Hosts per
network we can have. By determining the of bits
in the HostID portion of the address and figuring
out how many combinations of 1s and 0s there
could possibly be will tell us how many HostIDs
are available to us.
For example, with a default class C address 8
bits are available for the HostID (remember the
default Class C Subnet Mask!)
81
How many possible combinations of 8 bits are
there??
00000000, 00000001, 00000010, 00000011, ,
11111110, 11111111
Using the following formula is much easier than
counting!
Number of HostIDs 2n-2 where n bits
available for the HostID (or the of 0s in the
subnet mask)
subtract 2 because a Host ID of all 0s or all
1s is not allowed
82
The following table summarizes the number of
hosts available to each default IP Class
216 2 65,534
83
Quite often an ISP will divide up a Class C
address into smaller segments, since rarely does
a client need the full 254 addresses a Class C
address comes with. The ISP will divide up the
Class C address by subnetting it.
Subnetting is a way of manipulating the Subnet
Mask and dividing up an IP address into smaller
more manageable NetIDs.
84
Subnetting Example
Say an ISP had two customers who wanted to
connect their networks wirelessly to the
Internet. Each customer had 20 computers on
their networks and werent planning on adding any
more computers in the future.
85
The ISP has the NetID of 200.100.10.0 which it
can distribute to customers. How do we divide
this up into two segments for distribution to the
customer??
By default 8 bits are assigned to the HostID. If
we take bits away from the HostID and allocate
them to the NetID, it will give us more bits to
create more NetIDs.
86
Subnetting involves modifying the Subnet Mask and
taking bits away from the HostID to allow for
more Network IDs. In our example we need at least
2 NetIDs. In order to create 2 NetIDs we will
borrow 1 bit from the HostID portion of the
subnet mask.
Before Subnetting IP 200.100.10.0
11001000.01100100.00001010.00000000 SM255.255.255
.0 11111111.11111111.11111111.00000000
After Subnetting IP 200.100.10.0
11001000.01100100.00001010.00000000 SM255.255.255
.128 11111111.11111111.11111111.10000000
87
Basically, the more bits you borrow, the more
NetIDs you can create, but the fewer HostIDs per
network you are allowed. Use the following
formula to determine the of bits you need to
borrow
of new NetIDs 2n where n bits borrowed
from the HostID
88
Class C Subnetting
89
In our example we need to borrow at least one bit
and possibly up to 3 bits. If we borrow more
than three bits we will not have enough hosts per
network to support the 20 hosts on each network.
Lets assume we borrow 3 bits. That will leave us
3 bits available for the NetIDs
.00100000, .01000000, .01100000, .10100000,
.11000000, .11100000
and 5 bits available for the HostID in each
network.
90
What are the new NetIDs??
The easiest way to calculate the NetID of each
new network after subnetting is to take the
decimal value of the last borrowed bit and add it
to the original NetID.
Original NetID 200.100.10.0
New Subnet Mask 255.255.255.224
11111111.11111111.11111111.11100000
Decimal Value 32
91
Fourth Octet .00000000
Original NetID 200.100.10.0
32 200.100.10.32
.00100000
32 200.100.10.64
.01000000
32 200.100.10.96
.01100000
32 200.100.10.128
.10000000
32 200.100.10.160
.10100000
32 200.100.10.192
.11000000
32 200.100.10.224
.11100000
That leaves us with 5 bits for HostIDs in each
network.
92
The New Networks
200.100.10.0 with HostIDs from 1-30 200.100.10.32
with HostIDs from 33 - 62 200.100.10.64 with
HostIDs from 65 - 94 200.100.10.96 with HostIDs
from 97 - 126 200.100.10.128 with HostIDs from
129 - 156 200.100.10.160 with HostIDs from 161 -
190 200.100.10.192 with HostIDs from 193 -
222 200.100.10.224 with HostIDs from 224 - 253
93
HostIDs
The main rule to remember with the HostID is
that, in binary, a HostID of all 1s or all 0s is
invalidhence the reason for the 2 in our
formula 2n 2.
If we look at our first NetID of 200.100.10.0 and
the Subnet Mask of 255.255.255.224, when the
HostID is all 0s we have the equivalent of the
NetID
200.100.10.0
and with a HostID of all 1s we have the
broadcast IP address for the network.
200.100.10.31
94
This same theory holds true for all the NetIDs
generated by subnetting. The main thing to
remember about subnetting is that the more bits
you borrow, the more NetIDs you can create, but
the fewer hosts per subnet you will be
allowed. Use the following checklist to help
simplify subnetting
95
Subnetting Checklist

• Determine the number of NetIDs required currently
  and in the future.

• Determine the maximum number of hosts required
  per subnet currently and in the future.

• Define one Subnet Mask for the entire network
  that allows for the desired NetIDs and HostIDs.

4. Determine the resulting NetIDs that will be
used.
5. Determine the resulting HostIDs that will be
used.
96
Our New Network
NetID 200.100.10.0 SM 255.255.255.224
NetID 200.100.10.32 SM 255.255.255.224
NetID 200.100.10.64 SM 255.255.255.224
97
Subnet Mask Interpretation
Quite often different terminology is used to
identify the subnet mask. Instead of displaying
the entire subnet mask, an integer is often used
to identify how many bits are in use (ie. How
many 1s are in the subnet mask).
255.255.255.0 24 255.255.255.224
27 255.255.192.0 18
WaveRider products allow the use of either
representation when configuring the unit.
98
In this situation our NCL product will need to be
installed in Route mode, since we are connecting
different networks together. A general Rule of
Thumb, is to supply your networks Router with
the first available IP address in your IP class.
Our new network would look similar to this
99
2
1
3
100
Routers
Routers connect different networks together. A
router operates at the Internet layer of the TCP
Model. It uses IP addresses to determine how to
route data from one network to another. The
router uses its routing table to figure out where
to send the data destined for various networks.
The IP packet contains the source and destination
addresses for the packet being sent.
101
Since routers connect different networks together
we can conclude the following

• A router needs at least two interfacesone to
  connect each interface.

• A router needs at least two IP addressesone for
  each interface.

• A router needs a populated route table to know
  how to direct traffic.

102
WaveRider Routers

• When acting as a router, the WaveRider product is
  connected to two networks
• The ethernet network
• The radio network

Therefore, in our network we need to assign the
radio segment of our radios an IP address also.
For our example we will use 192.168.10.0 / 24
103
2
1
3
We now have 4 different IP networks.
104
Default Route Entries
By default a router has entries for the networks
to which it is attached as well as an entry for
the loopback address.
The Default Route Table for the 200.100.10.1
router would look like
A similar entry would appear in the other routers
as well
105
Establishing Route Tables

• The Routers do not know of other networks by
  default and must have their Route Tables built in
  order to forward traffic properly.
• Two methods for achieving this
• Static Route entries
• Dynamic Route entries

106
Static Routing
A static route is entered manually through the
command line interface. To add a route so that
packets could travel from 1 to 2 we would add
the following to 200.100.10.1
The entry is basically specifying what interface
to send data to when destined for a specific
NetID.
107
Data can now get from 1 to 2, but until we add
a route from 2 to 1, the communication will
only be in one direction.
2
1
ip route add 200.100.10.0 192.168.10.1 27
108
In order for 1 to send data to 3, and vice
versa, similar entries at the 200.100.10.1 and
200.100.10.65 routers will have to be made.
ip route add 200.100.10.64 192.168.10.3 27 ip
route add 200.100.10.0 192.168.10.1 27
109
2
1
3
We now have data flow from 1 to 2 1 to 3 3
to 1 2 to 1
110
Finally we need to add entries so that 2 can
talk to 3. The two most important rules to
remember are

• You can only pass data as far as the next router.
• You can only pass RF traffic to a radio which you
  have Line of Sight with.

Therefore, in our network for 2 to talk to 3,
all traffic must be passed though the router at
200.100.10.1
111
Eventually the route tables in all routers would
look like this
1
2
3
112
With both the NCL and the LMS radios, in a routed
network all traffic is passed through the
Master radio. Because of this, the route tables
at the Stations can be simplified with the
following entry
ip route add 0.0.0.0 ltmaster radiogt 0
This entry basically tells the local router to
forward any packet that it does not have a route
entry for to the Master radio. Since the Master
has a route entry to all other networks, the data
will still get to the intended destination.
113
The route tables will now look like this
1
2
3
114
Dynamic Routing
With Dynamic Routing there is no need for the
Network Administrator to manually add route
entries. Using the Router Information Protocol
(RIP) the routers broadcast their route tables
onto the network. Over time all routers will
learn routes to all other networks via these
broadcasts.
115
The LMS and the NCL product line both incorporate
RIP v2 into the product.
116
Network Troubleshooting Basics
117
Verifying Communications
The two most common ways of verifying that two
nodes on your network are communicating are
through the PING and the TRACERT command. Both
commands are run from the DOS window on a Windows
machine and allow you to verify whether or not
another IP address is accessible over the
network. They are both very handy in
troubleshooting by allowing you to determine
which link in the chain is causing problems.
118
PING uses ICMP Echo Request and Reply messages to
connect to another machine. If the other machine
is accessible it will reply to your request. If
a reply is not sent then usually there is either
a routing problem or an IP problem.
119
TRACERT is used to view all the hops that are
taken when passing through routers to a given
destination. It is useful in determining exactly
which hop is causing the communication problems.
120
What is the 1 cause of most Networking Problems??
BAD CABLING!!!
121
Twisted Pair Cabling
Most common type of cabling. Twisting of the
wires provides protection from crosstalk. UTP
commonly referred to as CAT3 or CAT5. The number
helps determine the speed of the cable. (CAT3
10Mbps CAT5 100Mbps). Supports cable runs up
to 100 meters (328 feet). Use RJ-45
connectors. STP contains shielding which makes it
less vulnerable to outside interference, but much
harder to install than UTP. AKA 10BaseT or
100BaseT
122
Use a crossed over UTP cable when connecting a
node to another node without using a hub, or when
connecting 2 hubs together that do not have an
uplink port.
123
Use a straight through UTP cable when attaching
nodes to a hub/switch or when attaching two hubs
using the uplink port.
124
Review Test