TCOM 509:
TCP/IP - Internet Protocols
Instructor: Scott T. Tran
* Obtained permission to use Raj Jain’s technical material
Course Objectives
Understand concept of datagram processing and
delivery (layering and encapsulation)
 Understand the client-server model as applied to
networking
 Understand IP Addressing and Subnet Masking
Schemes (CIDR/VLSM)
 Understand IP routing (RIP, OSPF, IS-IS)
 Understand service (e.g., application) addressing
and access to services across an IP network
 Understand TCP performance parameters and
metrics
 Advanced topics (IP Multicast, IP Tunneling, NAT,
DHCP, IP Security, etc…)

2
On-Line Course Info
 Look
week

at the website at least once per
http://osf1.gmu.edu/~stran4
3
IP Header
4
UDP Header
5
TCP Header
6
Communications Between LAN Hosts
(TCP/IP) Via Wide Area Networks (IP)
7
Chapter 1:
Introduction
Concept of Peer Entities – Logical
Relationships
9
Protocols
A
protocol is a set of rules and formats
that govern the communication between
communicating peers
 set
of valid messages
 meaning of each message
A
protocol is necessary for any function
that requires cooperation between peers
10
What does a protocol tell us?
 Syntax
of a message
 what
fields does it contain?
 in what format?
 Semantics
of a message
 what
does a message mean?
 for example, not-OK message means receiver got a
corrupted file
 Actions
to take on receipt of a message
 for
example, on receiving not-OK message, retransmit
the entire file
11
The Internet
Standards-based – the TCP/IP protocol suite
 An Open System – not linked to a single
vendor
 US Gov’t research resulted in quite and
extensible set of protocols






Best spent tax money I know of 
Evolution from gov’t-orientation to researchorientation to business-orientation
Why is it so good?
Why did it beat out “OSI Networking”?
Let’s start in on the details…
12
Getting Data from Here to There
IP is a form of packet switching
 Data is broken up into discrete chucks and then sent toward
destination
 Each packet has to find its own route to the destination. There is
no predetermined path; the decision as to which node to hop to in
the next step is taken only when a node is reached.
 Each packet finds its way using the information it carries, such as
the source and destination IP addresses.
 Network resources (routers, links) are shared between different
data streams - multiplexing
 The phone network: circuit switching
 Sender calls receiver and establishes a logical connection
 The connection is maintained for the duration of the data flow
 Two distinct paradigms
 Both have value
 The TCP part of TCP/IP provides a logical connection, when
necessary

13
Motivation behind OSI Model
 Is
a conceptual, reference model.
 Is
the primary architectural model for
inter-computer communications.
 Is
the only common language spoken by
different manufacturers.
 Mastery
of the OSI model is mandatory
14
Motivation behind OSI Model

The goal of an OSI layer is to communicate with its peer layer on
another host.
7
6
5
4
3
2
1

Application
Presentation
Session
Transport
Network
Data Link
Physical
7
6
5
4
3
2
1
Application
Presentation
Session
Transport
Network
Data Link
Physical
The information exchanged is called a Protocol Data Unit (PDU).
7
6
5
4
3
2
1
Application
Presentation
Session
Transport
Network
Data Link
Physical
Messages
Messages
Messages
Segments
Packets or Datagrams
Frames
Bits
15
Communication between OSI Layers
7
6
5
4
3
2
1
Application
Presentation
Session
Transport
Network
Data Link
Physical
7
6
5
4
3
2
1
Application
Presentation
Session
Transport
Network
Data Link
Physical
16
Motivation behind OSI Model – Protocol
Functionality










Fragmentation and Reassembly
 Breaking up data into pre-defined sized blocks
Encapsulation
 Adding control info to data (address, error detection code, etc…)
Connection Control
 Providing connection establishment, data transfer, connection termination
Flow Control
 Throttling of data rate exchanged between source and destination
Error Control
 Error detection
Synchronization
 Timeouts, Send state, Receive state. Etc…
Sequencing
 Numbering of data blocks (applicable only for connection-oriented mode)
Addressing
 Has local and global significance,
 Used for routing purposes in IP
Multiplexing
 Allowing multiple logical connections to use one physical connections
 Mapping of connections from one protocol layer to another
Transmission Services
 Security, Priority, Grade of Service
17
The OSI Reference Model
18
Why seven layers?






Need a top and a bottom -- 2
Need to hide physical link, so need datalink -- 3
Need both end-to-end and hop-by-hop actions; so
need at least the transport (TCP) and network
(IP) layers -- 5
Session and presentation layers are not so
important, and are often ignored
So, we need at least 5, and 7 seems to be
excessive
Note that we can place functions in different
layers
19
Physical layer


Moves bits between physically connected endsystems
Standard prescribes





Postal network


coding scheme to represent a bit
shapes and sizes of connectors
bit-level synchronization
Supported transmission: electric voltages, radio frequencies,
pulses of infrared or ordinary light
technology for moving letters from one point to another
(trains, planes, vans, bicycles, ships…)
Internet

technology to move bits on a wire, wireless link, satellite
channel etc.
20
Datalink layer

Introduces the notion of a frame




Idle markers tell us that a link is not carrying a
frame
Begin and end markers delimit a frame
On a broadcast link (such as Ethernet)





set of bits that belong together
end-system must receive only bits meant for it
need datalink-layer address
also need to decide who gets to speak next
these functions are provided by Medium Access sublayer
(MAC)
Some data links also retransmit corrupted
packets and pace the rate at which frames are
placed on a link


part of logical link control sublayer
layered over MAC sublayer
21
Datalink layer (contd.)
Datalink layer protocols are the first layer of
software
 Very dependent on underlying physical link
properties
 Usually bundle both physical and datalink layer
on host adaptor card



Postal service


example: Ethernet
mail bag ‘frames’ letters
Internet



a variety of datalink layer protocols
most common is Ethernet
others are FDDI, SONET, HDLC
22
Network layer
Logically concatenates a set of links to form
the abstraction of an end-to-end link
 Allows an end-system to communicate with
any other end-system by computing a route
between them
 Hides idiosyncrasies of datalink layer
 Provides unique network-wide addresses
 Found both in end-systems and in
intermediate systems
 At end-systems primarily hide details of
datalink layer



segmentation and reassembly
error detection
23
Network layer (contd.)
 At
intermediate systems
 participates
in routing protocol to create routing
tables
 responsible for forwarding packets
 scheduling the transmission order of packets
 choosing which packets to drop
24
Two types of network layers

In datagram (connection-less) networks
 provides both routing and data forwarding
 Ex: Internet using IP protocol

In connection-oriented networks
 we distinguish between data plane and control plane
 data plane only forwards and schedules data (touches every
byte)
 control plane responsible for routing, call-establishment, callteardown (doesn’t touch data bytes)
 Ex: TCP protocol running over IP
25
Network layer

Postal network





set up internal routing tables
forward letters from source to destination
static routing
multiple qualities of service
Internet






network layer is provided by Internet Protocol
found in all end-systems and intermediate systems
segmentation and reassembly
packet-forwarding, routing, scheduling
unique IP addresses
can be layered over anything, but only best-effort service
26
Transport layer - TCP



Network provides a ‘raw’ end-to-end service
Transport layer creates the abstraction of an
error-controlled, flow-controlled and
multiplexed end-to-end link
Error control



message will reach destination despite packet loss,
corruption and duplication
retransmit lost packets; detect, discard, and retransmit
corrupted packets; detect and discard duplicated packets
Flow control

match transmission rat to rate currently sustainable on the
path to destination, and at the destination itself
27
Transport Layer - TCP
28
Transport Layer (TCP) – Relationships with
Other Layers Below
Process-to-process delivery
29
Transport Layer Addressing
Addresses
•Data link layer  MAC address
•Network layer  IP address
•Transport layer  Port number (choose among multiple
processes running on destination host)
30
Transport layer (contd.)

Multiplexes multiple applications to the same
end-to-end connection


adds an application-specific identifier (port number)
so that receiving end-system can hand in incoming
packet to the correct application
Some transport layers provide fewer services


e.g. simple error detection, no flow control, and no
retransmission
lightweight transport layer
31
Transport layer (contd.)
 Postal
system
 doesn’t
have a transport layer
 implemented, if at all, by customers
 detect lost letters (how?) and retransmit
them
 Internet
 two
popular protocols are TCP and UDP
 TCP provides error control, flow control,
multiplexing
 UDP provides only multiplexing
32
Session layer



Not common
Provides full-duplex service, expedited data
delivery, and session synchronization
Duplex


Expedited data delivery


if transport layer is simplex, concatenates two
transport endpoints together
allows some messages to skip ahead in end-system
queues, by using a separate low-delay transport
layer endpoint
Synchronization

allows users to place marks in data stream and to roll
back to a pre-specified mark
33
Example

Postal network





suppose a company has separate shipping and
receiving clerks
chief clerk can manage both to provide abstraction of
a duplex service
chief clerk may also send some messages using a
courier (expedited service)
chief clerk can arrange to have a set of messages
either delivered all at once, or not at all
Internet

doesn’t have a standard session layer
34
Presentation layer


Unlike other layers which deal with headers,
presentation layer touches the application
data
Hides data representation differences between
applications




Can also encrypt data
Usually ad hoc
Postal network


e.g. endian-ness
translator translates contents before giving it to chief
clerk
Internet


no standard presentation layer
only defines network byte order for 2- and 4-byte
integers
35
Application layer



The set of applications that use the network
Doesn’t provide services to any other layer
Postal network








the person who uses the postal system
suppose manager wants to send a set of recall letters
translator translates letters going abroad
chief clerk sends some priority mail, and some by
regular mail
mail clerk sends a message, retransmits if not acked
postal system computes a route and forwards the
letters
datalink layer: letters carried by planes, trains,
automobiles
physical layer: the letter itself
36
OSI Reference Model vs. TCP/IP
Protocol Stack
OSI
TCP / IP
Application (Layer7)
Presentation (Layer6)
Application
Session (Layer 5)
Transport (Layer 4)
Transport
Network (Layer 3)
Internet
Data Link (Layer 2)
Physical (Layer 1)
Subnet
37
Layering and Encapsulation
38
Why Layering Required For Networking?

A system that is too complex to
comprehend in its entirety.

A system that is difficult to maintain.

A system whose least stable elements are
not isolated.

A system whose most reusable elements
are difficult to identify.

A system that is to be built by different
teams, possibly with different skills.
39
Why Need Encapsulation?
 Preserve
content of layers which is
private but allow interfacing between
them
 Allows
management of complexity and
change within layers
40
41
42
What is a Router?
A
specialized computer
 Interconnects multiple physical
networks
 Allows construction of a LOGICAL
network topology that is independent
from the PHYSICAL networks
 Notation
 Show-and-tell
43
How does a router's forwarding logic
differ from a bridge's forwarding logic?
1.
Packet Filtering: A router examines only those data packets specifically addressed to it, as opposed to a bridge,
which reads the destination address of every data packet on the LAN to which it is attached.
2.
Route Determination: A bridge checks the frame's data-link protocol for source as well as destination
address. It then checks its table of known local nodes. The destination address is compared with the contents
of the known local nodes in order to determine whether the frame should be allowed to cross the bridge or not
whether the destination is local or not). The bridge does not determine the path; it merely allows or disallows
the packet to cross. Destination routes must be obtained through other network devices, such as the
originating workstation for source routing bridges. Bridges are thus sometimes known as "forward if not
local" devices.
Routers actually maintain dynamic tables of "best routes", which depend on network conditions. Based the
latest traffic conditions, the router chooses the best path for the data packet to reach its destination, and
sends the data packet on its way. After reading the network layer destination address and the protocol of the
network layer data, the router consults its routing tables in order to determine the best path on which to
forward this data packet. Having found the best path, the router has the ability to repackage the data packet
as required for the chosen delivery route. For example, if the packet were to be sent out over an X.25 packetswitched network, the router would encapsulate the packet in an X.25-compliant envelope.
3. Routing Logic: A bridge reads the destination address of each data frame on a LAN, decides whether the
address is local or remote (on the other side of the bridge), and only allows those data frames with non-local
destination addresses to cross the bridge. A router is more discriminating. The router first confirms the
existence of the destination address as well as the latest information on available network paths to reach that
destination. Unlike a bridge, which merely allows access to the internetwork (forward-if-not-local logic), a
router specifically addresses the data packet to a distant router. However, before a router actually releases a
data packet on to the internetwork, is confirms the existence of the destination address to which the data
packet is bound. Only once the router is satisfies with the viability of the destination address as well as with
the quality of the intended path, will it release the packaged packet. The router's meticulous processing is
known as "forward if proven remote" logic.
44
TCP/IP Protocol Suite Layers
45
TCP/IP Layering
46
Multiplexing and Demultiplexing
47
The Client-Server Model
 Two
hosts interact in a predefined
manner
 One
side is the client – it wants information
 One side is the server – it provides the
information
 EX: WWW – web browser is the client,
website is the server
A
host is not “locked down” to be only a
client or only a server
 Multiple
client and server programs can be
running on a single host at the same time
48
Client/Server on the same LAN
49
Client/Server on two different LANs
50
Service Addresses: Port Numbering
 Port
numbers are “Layer 4” addresses
(TCP or UDP in the TCP/IP suite)
 They allow multiple services on a single
host to have unique addresses
 E.g.,
one host can be running servers for
FTP, HTTP, and telnet
 Each service listens on it’s own port
 The
combination of IP address plus
TCP/UDP can uniquely identify a
connection (a “socket”)
51
How Do You Get a Port Number?
Standards and Standards Processes


The Internet Engineering Task Force (IETF)
most directly controls the development of
standards for the TCP/IP protocol suite
Those standards are called Requests for
Comment (RFCs)


Relatively collegial process
Different from IEEE, ITU, ANSI, etc.
http://www.ietf.org
 http://www.rfc-editor.org
 See section 1.11, pp. 14-15


Some of these RFCs have been updated
52
IP Protocol and Its Associated Upper
Layer: IP Protocol Numbers
http://www.iana.org/assignments/protocol-numbers
unix-host% more /etc/protocols
#
# Internet protocols
#
# $FreeBSD: src/etc/protocols,v 1.13.2.1 2000/09/24 11:26:39 asmodai Exp $
#
from: @(#)protocols
5.1 (Berkeley) 4/17/89
#
# See also http://www.isi.edu/in-notes/iana/assignments/protocol-numbers
#
ip
0
IP
# internet protocol, pseudo protocol number
#hopopt 0
HOPOPT
# hop-by-hop options for ipv6
icmp
1
ICMP
# internet control message protocol
igmp
2
IGMP
# internet group management protocol
ggp
3
GGP
# gateway-gateway protocol
ipencap 4
IP-ENCAP
# IP encapsulated in IP (officially ``IP'')
st2
5
ST2
# ST2 datagram mode (RFC 1819)
tcp
6
TCP
# transmission control protocol
cbt
7
CBT
# CBT, Tony Ballardie <A.Ballardie@cs.ucl.ac.uk>
egp
8
EGP
# exterior gateway protocol
igp
9
IGP
# any private interior gateway (Cisco: for IGRP)
<snip>
udp
17
UDP
# user datagram protocol
ipv6
41
IPV6
# ipv6
sdrp
42
SDRP
# Source Demand Routing Protocol
ipv6-route
43
IPV6-ROUTE
# routing header for ipv6
ipv6-frag
44
IPV6-FRAG
# fragment header for ipv6
idrp
45
IDRP
# Inter-Domain Routing Protocol
rsvp
46
RSVP
# Resource ReSerVation Protocol
gre
47
GRE
# Generic Routing Encapsulation
<etc.>
53
Chapter 2:
Link Layer
Ethernet Encapsulations – 2 Types
55
Loopback Interfaces
Special logical IP address (127.0.0.1)
 Any IP traffic sent to loopback interface must not appear on any network
 Used to allow a client and a server on the same host to communicate with
each other using TCP/IP

56
Maximum Transmission Unit (MTU)
57
Chapter 3:
IP: Internet Protocol
IP Header
59
60
61
62
IP Fragmentation & Reassembly
network links have MTU
(max.transfer size) - largest
possible link-level frame.
 different link types,
different MTUs
 large IP datagram divided
(“fragmented”) within net
 one datagram becomes
several datagrams
 “reassembled” only at
final destination
 IP header bits used to
identify, order related
fragments

fragmentation:
in: one large datagram
out: 3 smaller datagrams
reassembly
63
IP Fragmentation and Reassembly






IP header has
identification (x), flag,
and fragmentation
fields
Example: 4000byte
d’gram (20byte header
+ 3980 IP payload).
MTU = 1500bytes
Frag 1: 1480bytes +
20byte header
Frag 2: 1480bytes +
20byte header
Frag 3: 3980-2*1480
bytes + 20byte header
length ID fragflag offset
=4000 =x
=0
=0
One large datagram becomes
several smaller datagrams
length ID fragflag offset
=1500 =x
=1
=0
length ID fragflag offset
=1500 =x
=1
=1480
length ID fragflag offset
=1040 =x
=0
=2960
64
IP Fragmentation processing at a Router

Fragmentation is performed when packet size is larger than MTU size
of the outgoing interface

To fragment/segment a long internet packet, an Intermediate System
using the Internet Protocol (for example, a router), creates two new
IP packets and copies the contents of the IP header fields from the
long packet into BOTH new IP headers.

The data of the long packet is divided into two portions on a 8 byte
(64 bit) boundary. All packets which have a more fragments (MF) flag
set, must have an integral multiple of 8 bytes, but those that do not
have this flag set need not do.

If we call the number of 8 byte blocks in the first portion NFB (for
Number of Fragment Blocks). The first portion of the data is placed in
the first new IP packet, and the total length field is set to the length
of the FIRST IP packet. The more-fragments flag (MF) is set to one.

The second portion of the data is placed in the second new IP packet,
and the total length field is set to the length of the SECOND packet.
The more-fragments flag (MF) carries the same value as the long
packet. The fragment offset field of the second new IP is set to the
value of that field in the long IP packet plus the NFB.
65
IP Addresses in TCP/IP Model
Addresses provide UNIQUENESS
 Different from other types of address





Layer 1 – physical address: hardware manufacturer
assigns, hardly ever changes; MAC addresses
Layer 2 – logical address: you assign, and reassign as
changes are made in the network; IP addresses
Layer 3 – “service” address: standards bodies assign,
software manufacturers must abide by them for
interoperability; TCP/UDP addresses
Why 32 bits for IP?

2^32 = 4+ billion – enough addresses (???)
66
Addressing, Numbering, and
Notation

Computers care about BINARY


Humans care about DECIMAL





IP addresses are truly 32-bit unsigned integers,
represented in dotted-decimal (a.k.a. dotted-quad)
for our convenience
Electrical Engineers care about HEXADECIMAL


On/off, hi/low, signal/no signal, etc.
Neither computer nor human? 
Compact representation of binary info
Often used for Layer 2 (hardware) addresses
YOU need to care about ALL 3
Discuss some examples here…
67
Bit Positions and Their Values
68
IP Address Classes
69
IP Address Class Ranges
70
Address Class Characteristics
Class
Network Bits
Host Bits
Total Networks
Total Addresses
A
8
24
127
16,777,216
B
16
16
16,384
65,536
C
24
8
2,097,152 256
71
IP addressing: the last word...
Q: How does an ISP get block of addresses?
A: ICANN: Internet Corporation for Assigned
Names and Numbers (guidelines in RFC 2050)
 allocates addresses
 manages DNS
 assigns domain names, resolves disputes
72
Subnetting 
To divide the standard classful host-number field into two parts - the
subnet-number and the host-number on that subnet.

Motivation:
 Efficient use of available network addresses
 Flexibility in planning network growth and design
 Capability to contain broadcast traffic (ARP, RARP, etc…)
 Subnets under local administrative control

Mechanism:
 Define/assign a subnet mask for addresses in a network that has been subnetted
 Subnet mask tells router which octets of an IP address to pay attention to
when comparing the destination address of a packet to its routing table
entries
 A subnet mask identifies the subnet field of network addresses
 Correct routing requires that all subnets of a network be physically
contiguous. In other words, the network must be set up such that it does
not require traffic between any two subnets to cross another network
 Most implementations require that all subnets of a network have the same
number of subnet bits.

Example
73
Arrangement of Subnets
74
How To Calculate Subnet Address with a Given
Subnet Mask
Resulting subnet address is 171.16.1.0
75
Example Subnet Masks
76
Using Subnet Masks
77
Supernetting – CIDR and VLSM

Motivation: Address issues with current IP Address Depletion

Subnetting allows you to take a (classful) block of
addresses and break it up into usable portions
 Subnetting >>> Segregation

Supernetting allows you to implement classless
addressing scheme and combine address blocks for the
purposes of efficiency in routing updates
 Supernetting >>> Aggregation
 Rationale: More flexible use of IP addresses and
reduces entries in the routing table

Two ways to implement Supernetting
 Organizations need Variable Length Subnet Mask
(VLSM) to provide flexibility and address efficiency
 The Internet needs Classless Interdomain Routing
(CIDR) for scalability
78
Requirements for Deploying CIDR or
VLSM
 The
successful deployment of VLSM has three
prerequisites:
 The
routing protocols must carry extended network
prefix information with each route advertisement.
 All
routers must implement a consistent forwarding
algorithm based on the “longest match.”
 For
route aggregation to occur, addresses must be
assigned so that they have topological significance
79
Comparing CIDR to VLSM

CIDR and VLSM both allow a portion of the IP address
space to be recursively divided into subsequently smaller
pieces. The difference is that with VLSM, the recursion is
performed on the address space previously assigned to
an organization and is invisible to the global Internet.

CIDR, on the other hand, permits the recursive allocation
of an address block by an Internet Registry to a highlevel ISP, a mid-level ISP, a low level ISP, and a private
organization’s network.
80
Classless Interdomain Routing (CIDR)

With subnet addressing, we can have higher flexibility
within a domain/AS

However, the rigidity of classful IP addresses is still very
inflexible (e.g. HostIDs in a domain is limited to 256, 66048,
16908288)

CIDR – use arbitrary prefix length of Network ID


E.g. 205.100.0.0/22 means that network ID length is 22 bits,
i.e. netmask is 255.255.252.0
Also allows RECURSION allocation of an address block
provided by the Internet Registry to a high-level ISP, to a
mid-level ISP, to a low-level ISP, and finally to a private
organization’s network
81
Classless Interdomain Routing (CIDR)
Changes to routing table
 Each entry must specify a 32-bit mask together with the
32-bit IP address
 Use longest prefix match to find a suitable entry
 E.g. a packet with destination IP addr: 205.100.1.2, and the
routing has two entries as
205.100.0.0/22 and 205.100.0.0/20.
Both entries match the destination IP addr, which one should
be chosen? Ans: Choose the one with longest matched bits
 205.100.0.0/22
=
11001101.01100100.00000000.00000000
 205.100.0.0/20
=
11001101.01100100.00000000.00000000
 205.100.1.2
=
Longest match 11001101.01100100.00000001.00000010

82
Classless Interdomain Routing (CIDR)
 Advantages:
 An
organization can “buy” the number of IPs
according to its needs (not confined to 256,
66048, 16908288)
 Reduce routing table size significantly as
multiple “continuous” networks following the
same route can be combined to form a single
routing entry
 E.g.
original 4 entries for destinations as
 137.188.0.0, 137.189.0.0, 137.190.0.0,137.191.0.0

Now, we can combine them into one entry of
 137.188.0.0/14
83
An Example of How CIDR Is Used
CIDR Reduces the Size of Internet Routing Tables
84
CIDR: Partitioning of IP addresses
 Q:
How does network get network part of
IP addr?
 A: gets allocated portion of its provider
ISP’s address space
ISP's block
11001000 00010111 00010000 00000000
(allocated to ISP). It is divided into 8 equal sized blocks.
Organization 0 11001000 00010111 00010000 00000000
Organization 1 11001000 00010111 00010010 00000000
Organization 2 11001000 00010111 00010100 00000000
...
…..
….
200.23.16.0/23
200.23.18.0/23
200.23.20.0/23
….
Organization 7
200.23.30.0/23
11001000 00010111 00011110 00000000
200.23.16.0/20
85
CIDR Hierarchical addressing: route
aggregation
Hierarchical addressing allows efficient advertisement of routing
information: “Fly-by-night-ISP requests that all datagrams whose first
20 address bits match 200.23.16.0/20. The world doesn’t know that
within this there are 8 other orgs. each with their own networks.
Organization 0
200.23.16.0/23
Organization 1
200.23.18.0/23
Organization 2
200.23.20.0/23
Organization 7
.
.
.
.
.
.
Fly-By-Night-ISP
“Send me anything
with addresses
beginning
200.23.16.0/20”
Internet
200.23.30.0/23
ISPs-R-Us
“Send me anything
with addresses
beginning
199.31.0.0/16”
86
Hierarchical addressing: more specific
routes
Suppose Org. 1 dislikes Fly-by-night-ISP’s service and wants to move to
ISPs-R-Us? Org.1 keeps its addresses in 200.23.18.0/23 but now
ISPs-R-Us advertises 200.23.18.0/23.
When other routers see 200.23.16.0/20 &
200.23.18.0/23 and want to route to 200.23.18.0/23
They will use the longest prefix matching rule
and send to ISPs-R-Us
Organization 0
200.23.16.0/23
Organization 2
200.23.20.0/23
Organization 7
.
.
.
.
.
.
Fly-By-Night-ISP
“Send me anything
with addresses
beginning
200.23.16.0/20”
Internet
200.23.30.0/23
ISPs-R-Us
Organization 1
200.23.18.0/23
“Send me anything
with addresses
beginning 199.31.0.0/16
or 200.23.18.0/23”
87
Supernetting – VLSM

In 1987, RFC 1009 specified how a subnetted network could
use more than one subnet mask. When an IP network is
assigned more than one subnet mask, it is considered a
network with (VLSM) since the extended network prefixes
have different lengths.

Allows RECURSIVE division of a network prefix (subnets of
subnets)

Allows detailed structure of routing info for one subnet group
to be hidden from routers in another subnet group

VLSM is different than CIDR because the recursion is
performed on the address space previously assigned to an org.
and is INVISIBLE to the global Internet
88
VLSM Design Considerations

When developing a VLSM design, the network designer
must recursively ask the same set of questions as for a
traditional subnet design. The same set of design
decisions must be made at each level of the hierarchy:
1 How many total subnets does this level need today?
2 How many total subnets will this level need in the
future?
3 How many hosts are on this level’s largest subnet
today?
4 How many hosts will be on this level’s largest subnet
be in the future?
89
An Example of How VLSM Is Used
Recursive Division
Detailed Structure
Of Recursion is hidden
VLSM Permits Route Aggregation - Reducing Routing Table Size
90
IP Design Considerations
 Addressing
(Impact of Subnetting)
 Routing (Topology dependent)
 Fragmentation and reassembly (MTU
size for different layers)
 Datagram lifetime (impact of TTL
setting)
 Error control (Related to MTU size)
 Flow control (limited via ICMP)
91
IP Futures
 In
addition to Stevens’ observations:
 IP
version 6
 IPSec
 MPLS
 IP Multicast
 These aren’t really futures anymore; they’re
here today
 Think
about what you’d like to cover in
last week
 Time
permitting
 I’m leaning toward IPSec
92
IPv4 addressing - Summary
 There
are three types of IPv4 addressing
environments.
 Original
Classful.
 Classful and subnet mask (RFC 950).
 Classless.
 An
internetwork can be a mix of several
environments.
93
Chapter 4:
ARP: Address Resolution
Protocol
To ARP or Not to ARP?
That is the question.
TCP/IP Layering
95
Multiplexing and Demultiplexing
96
TCP/IP suite: ARP





Maps IPv4 addresses to MAC addresses.
An ARP request is a local broadcast.
ARP broadcasts are not propagated through
routers.
Entries in the ARP table are deleted when a
timeout expires.
Several scenarios exist:




You know the IP address to send to and you need the
corresponding MAC address (ARP)
You know your MAC address and you need to know your IP
address (RARP)
You’re hiding physical networks (Proxy ARP)
You need to verify that your IP address isn’t being used by
another host (Gratuitous ARP)
97
98
Packet Delivery on a single LAN
99
Use of ARP - Packet Delivery across
multiple LANs
100
The Purpose of ARP
101
An Example:
ARP with TFTP
102
ARP Frame Format
103
ARP Notes
 ARP
generally only occurs on a single
physical network
 ARP
request is a layer 2 broadcast, and
routers block these broadcasts by default
 ARP
is designed to work for protocols
other than IP
A
generic solution
 Some other protocols (e.g., IPX) were
designed such that they don’t need ARP
104
More ARP Notes
 Dynamic
nature of ARP is very flexible
 MAC
addresses tend to stay the same, but
IP addresses can change (e.g., DHCP,
change in logical structure of IP network)
 Sometimes MAC addresses can change (e.g.,
change a broken NIC, administratively
change MAC address)
 ARP allows for dynamic (re-)mapping
 What
happens if you ARP every time?
 Lots
of overhead
 Use a cache mechanism with timeouts
105
Notes on Proxy ARP

Also known as “promiscuous ARP”

Accommodates older TCP/IP stacks. A technique by which a router replies
to an Address Resolution Protocol (ARP) request from a host on behalf of
the ARP target host.

Proxy ARP (Address Resolution Protocol) is a technique by which a network
host answers to the ARP queries for the network address that it does not
have configured on the receiving interface. Proxying ARP requests on
behalf of another host effectively directs all LAN traffic destined for that
host to the proxying host/router. The "captured" traffic is then typically
routed to the destination host via another interface or via a tunnel.

When you see same MAC address in ARP cache for 2 different IP
addresses, that’s a hint that Proxy ARP is being used

Proxy ARP can create DoS attacks on networks if misconfigured. For
example a misconfigured router with proxy ARP has the ability to receive
packets destined for other hosts (as it gives its own MAC address in
response to ARP requests for other hosts/routers), but may not have the
ability to correctly forward these packets on to their final destination, thus
blackholing the traffic.
106
Uses of Gratuitous ARP

When a computer starts, a packet is broadcast on the network
containing the computer's TCP/IP address to prevent the use
of duplicate addresses on the same network

When a computer starts and its Ethernet hardware address
has changed due to interface card replacement, a packet is
broadcast to other host to signal an update to the IP-to-MAC
address mapping
107
Issues With ARP
 Weak
Security
a
bogus host can issue a gratuitous ARP and
change cache entries on other router’s
cache table
 a bogus host can send replies giving its own
hardware address (instead of the target) –
re-directing traffic
 Broadcasting
 excessive
 CPU
can be expensive
use of bandwidth
costs
108
Chapter 5:
RARP: Reverse Address
Resolution Protocol
What is RARP For?
A
workstation without a hard drive
(e.g., diskless workstation, X-terminal,
“thin client”) may have no means to
“remember” an IP address
 However, it will have a NIC that has a
MAC address burned-in on an EEPROM
 RARP allows this host to broadcast and
request it’s IP address
A
RARP Server must be configured on the
local subnet to assign this particular MAC
address with a unique IP address
110
Issues with RARP
 More
difficult to implement than ARP
 ARP
is needed for basic IP communications
and requires no configuration (mostly)
 RARP config normally resides in a static text
file
 Coordination between multiple RARP
servers requires that those text files are
always in sync
 Improvements
over RARP
 BOOTP
 DHCP
– most commonly used today
111
TCP/IP suite: DHCP vs. RARP
 RARP
is based on a table that needs to be
configured in the RARP server.
 Static, one-to-one address mapping: The
same MAC address will always acquire the
same IP address.
 RARP does provide IP addresses to
devices, but there is no much gain in
administrative overhead.
112
TCP/IP suite: DHCP vs. RARP
 RARP
is obsolete and almost never seen.
 DHCP is implemented in many devices
such as Windows NT servers, Novell
servers, Cisco routers, NAT boxes…
 Although there are plans for DHCPv6, IPv6
has some auto-configuration mechanisms
that will probably make DHCP obsolete in
the long run.
113
TCP/IP suite: DHCP
 This
reservation mechanism provides a
functionality similar to RARP: the IP
address is obtained from the server and is
always the same.
 A typical IP set is a combination of static,
reservations, and dynamic DHCP
addresses.
114
TCP/IP suite: DHCP
 DHCP
is an evolution of BOOTP
 Provides the same basic functionality as
RARP, but the underlying mechanism is
not the same.
 Can provide additional functionality such
as the address of the WINS server or the
node-type.
115
TCP/IP suite: DHCP vs. RARP
 Unless
a reservation is made, there is no
guarantee that a device will obtain the
same IP address each time.
 Servers are typically configured with static
IP addresses.
 DHCP does take some of the
administrative burden out (for
workstations).
116
DHCP: Dynamic Host Configuration Protocol
Goal: allow host to dynamically obtain its IP address
from network server when it joins network
Can renew its lease on address in use
Allows reuse of addresses (only hold address while
connected an “on”
Support for mobile users who want to join network
DHCP overview:
 host broadcasts “DHCP discover” msg
 DHCP server responds with “DHCP offer” msg
 host requests IP address: “DHCP request” msg
 DHCP server sends address: “DHCP ack” msg
117
DHCP client-server scenario
A
B
223.1.2.1
DHCP
server
223.1.1.1
223.1.1.2
223.1.1.4
223.1.2.9
223.1.2.2
223.1.1.3
223.1.3.1
223.1.3.27
223.1.3.2
E
arriving DHCP
client needs
address in this
network
118
TCP/IP suite: DHCP vs. RARP
 There
are some circumstances where the
use of DHCP would be nice, but the
devices needs a fixed IP address. Example:
Print Servers.
 The
DHCP solution is called reservation.
The MAC address of the device is
configured in the DHCP server.
119
Chapter 6:
ICMP: Internet Control
Message Protocol
Where ICMP in The TCP/IP Layering
121
What is ICMP?
 Every
protocol suite needs mechanisms
for control and error messaging
 Phone
network to end user: dial tone,
ringing tone, etc. (note: in-band)
 Phone network between switches: SS7
network for call management (note: out-ofband)
 ICMP
is the set of messages that handle
basic control and error messaging for
the TCP/IP protocol suite
122
ICMP Message Types
123
Key ICMP Uses
Time Stamp
 Ping
 Traceroute
 Source Quench




Indication that flow control needs to be activated at
the source
MTU size determination
ICMP Destination Unreachability

Can be used to detect malicious port scanning
activity
124
125
ICMP Message Transport in IP
126
ICMP Message Format
127
ICMP Timestamp
128
Round Trip Time (RTT) Concept
129
Chapter 7:
ping
What is ping?
 Uses
ICMP Echo Request and Reply
 Tests
reachability – make sure that the
network connection is in tact
 Don’t
use it for fine-grained
measurements of network performance
131
ICMP packets used for ping
132
133
Ping output
134
Ping with IP Options: Record Route
135
ping with Record Route
136
ICMP Source Quench

When to send an ICMP Source Quench



What to do when an ICMP Source Quench is
received



Standards says: when a packet is dropped inside a router
due to depleted buffer space
Real life: when ½ of the buffer space is used up
Implementation dependent
Ex: Reset the window size to 0 for n number of ACKS have
been received
Security Concern: Source Quench messages are
used by attackers in ICMP flood attacks
137
MTU Size Determination
138
ICMP Destination Unreachability

16 different categories of Destination Unreachable ICMP
messages

Ex: ‘port unreachable’ (type 3, code 3) where a local host requests
information from a remote host using TCP or UDP, and the remote host
doesn’t have an application listening on the required port. The remote
host replies with the type 3, code 3, ICMP messages declaring the
problem

Security Concern: These messages outbound will enable an attacker
the ability to easily map network topology
139
ICMP Port Unreachable – Example msg
140
ICMP Port Unreachable - Format
141
Chapter 8:
traceroute
IP Routing Processes

The IP routing processes on all nodes involved in the delivery of an IP packet includes: the sending host, the
intermediate routers, and the destination host.

IP on the Sending Host

When a packet is sent by a sending host, the packet is handed from an upper layer protocol (TCP, UDP, or ICMP) to IP. IP on the
sending host does the following:

Sets the Time-to-Live (TTL) value to either a default or application-specified value.

IP checks its routing table for the best route to the destination IP address.
If no route is found, IP indicates a routing error to the upper layer protocol (TCP, UDP, or ICMP).

Based on the most specific route, IP determines the forwarding IP address and the interface to be used for forwarding the
packet.

IP hands the packet, the forwarding IP address, and the interface to Address Resolution Protocol (ARP), and then ARP resolves
the forwarding IP address to its media access control (MAC) address and forwards the packet.

IP on the Router - When a packet is received at a router, the packet is passed to IP. IP on the router does the following:

IP verifies the IP header checksum.
If the IP header checksum fails, the IP packet is discarded without notification to the user. This is known as a silent discard .

IP verifies whether the destination IP address in the IP datagram corresponds to an IP address assigned to a router interface.
If so, the router processes the IP datagram as the destination host (see step 3 in the following "IP on the Destination Host"
section).

If the destination IP address is not the router, IP decreases the time-to-live (TTL) by 1.
If the TTL is 0, the router discards the packet and sends an ICMP Time Expired-TTL Expired message to the sender.

If the TTL is 1 or greater, IP updates the TTL field and calculates a new IP header checksum.

IP checks its routing table for the best route to the destination IP address in the IP datagram.
If no route is found, the router discards the packet and sends an ICMP Destination Unreachable-Network Unreachable message to
the sender.

Based on the best route found, IP determines the forwarding IP address and the interface to be used for forwarding the packet.

IP hands the packet, the forwarding IP address, and the interface to ARP, and then ARP forwards the packet to the appropriate
MAC address.

This entire process is repeated at each router in the path between the source and destination host.

IP on the Destination Host - When a packet is received at the destination host, it is passed up to IP. IP on the destination
host does the following:

IP verifies the IP header checksum.
If the IP header checksum fails, the IP packet is silently discarded.

IP verifies that the destination IP address in the IP datagram corresponds to an IP address assigned to the host.
If the destination IP address is not assigned to the host, the IP packet is silently discarded.

Based on the IP protocol field, IP passes the IP datagram without the IP header to the appropriate upper-level protocol.
If the protocol does not exist, ICMP sends a Destination Unreachable-Protocol Unreachable message back to the sender.

For TCP and UDP packets, the destination port is checked and the TCP segment or UDP header is processed.
If no application exists for the UDP port number, ICMP sends a Destination Unreachable-Port Unreachable message back to the
sender. If no application exists for the TCP port number, TCP sends a Connection Reset segment back to the sender.
143
What is traceroute?
A
program ported to just about anything
with a TCP/IP stack
 Shows the path packets take across the
network
 Takes
advantage of the ICMP “time
exceeded” message
 “tracert”
in Windows products
144
How Traceroute Works

Traceroute creates an UDP packet with the time to live (TTL) in the IP Header set to 1 and
addresses the packets set to the destination computer's IP address

Traceroute process waits for a response. This response will be:


An ICMP Time Exceeded message - this means the host responding is not the destination.
An ICMP Port Unreachable - this means the UDP layer at the destination host responding doesn't not
recognize the UDP port info in the received UDP packet.

The computer on which the messages die because the time to live expired (somewhere
between the Source and Destination hosts ) sends back ICMP Time Exceeded (ICMP Type
'11') responses. These messages indicate to the soure that the traceroute messages have not
yet reached the destination host

The source increments the TTL in the IP Header by one, then repeats steps the previous six
steps (creates 3 packets, sets the Time to Live to the next highest number, starts a timer,
transmits the packets, waits for a response). This process is repeated until the packets reach
the destination computer which the source host is tracing the route to.

When the ICMP message reaches the destination computer, the UDP layer will get to process
it and will find out that the UDP port specified is invalid which will trigger an ICMP Port
Unreachable message back to the source host.

The Port Unreachable error message indicates to traceroute that the destination has been
reached.
145
ICMP time exceeded message
146
How to read traceroute output
147
148