1McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Chapter 20

Network LayerProtocols:

ARP, IPv4, ICMPv4, IPv6, and ICMPv6

2McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Protocols at Network Layer

IP is responsible for host-to-host delivery of datagrams from source to destination.

ARP: Find the MAC (Physical) address of the next hop. Data link layer encapsulates this address into the frame

ICMP: Handle unusual situations such as the occurrence of an error.

IP is meant for unicast. For Multicast, we need IGMP.

3McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

ARP Hosts and routers are recognized at the

network level by their IP addresses. IP is unique.

At physical network level, we use MAC. MAC is unique locally but not necessarily universally.

We need both IP and MAC address because a physical network, such as Ethernet, can have two different protocols at the network layer, such as IP and IPX (Novell), at the same time. Likewise, a packet at a network layer such as IP may pass through different physical networks, such as Ethernet and Token Ring.

IP and MAC address need to be mapped.

4McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Static Mapping Having a table of IP and MAC address mapping in

all the machines. Limitations if MAC address changes due to

Change of network card In LocalTalk, MAC address changes when a machine is

turned on A mobile computer can move from one network to

another and so can gain different MAC address. Dynamic mapping

Each machine knows one of the two addresses. Use a protocol to get the other address. ARP & RARP. RARP is now replaced by DHCP. ARP associates an IP address with its MAC

address.

ARP mapping

5McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

ARP Operation

ARP request has the sender’s IP and MAC & the receiver’s IP address. It’s a broadcast as the physical address of the receiver is unknown.

All hosts in the network processes this request but only the intended recipient responds. Response is unicast.

6McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Hardware Type: 16-bits; Defining the type of network; Ethernet is 1.

Protocol Type: 16-bits; IPv4 is 0800. Hardware and Protocol length is 8-bits.

ARP Packet Format

7McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Encapsulation of ARP Packet

Start of Frame Delimiter -SDF

8McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Sender knows the IP address of the target. IP asks ARP to create ARP request message. It

includes sender’s and target’s IP and physical addresses. Target Physical address is all 0s.

Message is passed to data link layer, encapsulated in a frame using sender’s physical address. The physical destination address is the broadcast address.

All machines drop the packet except the targeted machine. Target machine identifies the IP address.

Target machine sends the ARP reply with its physical address.

Sender receives the reply and knows the physical address of target.

IP datagram, carries data for target machine, is now encapsulated in a frame and is unicast to the destination.

Delivery of the Datagram

9McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Four cases using ARP

10McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Case 1: Map the destination IP address to the physical address of the destination host (host MAC).

Case 2: Map the IP address of the router to the physical address (router MAC).

Case 3: IP of next router is mapped to the physical address (MAC of next router)

Case 4: Destination IP is mapped to the destination MAC

Four cases using ARP

11McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Example 1Example 1

A host with IP address 130.23.3.20 and physical address B23455102210 has a packet to send to another host with IP address 130.23.43.25 and physical address A46EF45983AB. The two hosts are on the same Ethernet network. Show the ARP request and reply packets encapsulated in Ethernet frames.

SolutionSolution

Figure 20.6 shows the ARP request and reply packets. Note that the ARP data field in this case is 28 bytes, and that the individual addresses do not fit in the 4-byte boundary. That is why we do not show the regular 4-byte boundaries for these addresses. Note that we use hexadecimal for every field except the IP addresses.

12McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Figure 20.6 Example 1

Start of Frame Delimiter -SDF

13McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Host-to-host network layer delivery protocol for the Internet.

Unreliable and connectionless datagram protocol

Best-effort: no error control or flow control. Has error detection mechanism to discard

the packets that are corrupted. For reliability, use IP with TCP. Each datagram is delivered independently

and via different routes. Datagrams: Packets of IP layer.

Variable-length packet consisting of header [20 to 60 bytes] and data.

IP

14McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

IP datagram

HLEN must be multiplied by 4 to get the length in bytes Differentiated services: QoS Total length = length of data + header length TTL: number of hops; approx 2 times the maximum number

of routes between any two hosts. Protocol: Defines the higher level protocol that uses IP

layer.

15McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Protocols: Multiplexing

Value Protocol

• 1 ICMP

• 2 IGMP

• 6 TCP

• 17 UDP

• 89 OSPF

16McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Checksum only covers the header and not data. Data checksum is handled by higher-level

protocols that encapsulate the data in the IP datagram.

Header changes when packets travels on the network but data does not change.

Calculate Checksum Divide the IP header into 16-bit sections. Value of checksum field is set to zero. All the sections are added and the sum is

complemented. The result is inserted in the checksum field.

Checksum

17McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Example of checksum calculation

18McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Fragmentation

Fragmentation Each router on the network decapsulates the IP datagram from the

frame it receives, process it, and then encapsulates it in another frame.

Format and size depends on the incoming and outgoing physical network.

IP datagram must be divided to make it possible to pass through these physical networks. This is called fragmentation

Maximum Transfer Unit (MTU) To make IP independent of the physical network, the maximum length

of IP is equal to the largest maximum transfer unit (MTU) 65,535 bytes.

19McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Identification: Same for all the fragments. Helps destination in reassembly of fragments.

Flags: 3-bit field. First bit is reserved. Second bit is called Don’t fragment bit. If set, don’t fragment the

packet. If set and still needs fragmentation, discard the packet and send an ICMP message to the source host.

Third field is more fragment bit. If 0 means that this is the last fragment.

Fragmentation offset 13-bit field shows the relative position of this fragment with respect

to the whole datagram. It is the offset of the data in the original datagram measured in

units of 8 bytes. Forces hosts or routers that fragment datagrams to choose the size

of each fragment so that the first byte number is divisible by 8.

Fragmentation Fields

20McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Fragmentation Example

If needed, the packets are fragmented. Each fragmented datagram has a header of its own. A fragmented datagram may itself be fragmented if it encounters a network with an even smaller MTU.

Fragmentation is done at source or at the routers on the fly. Re-assembly is done only at the destination.

21McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

ICMP: Internet Control Message Protocol

IP lacks error control. No error-reporting or checking. What happens if router can not find a final destination? What if time-to-live filed has zero value? What will happen if final destination has not received all

the fragments within a pre-determined time limit? Lacks of assistance mechanisms.

A host sometimes needs to determine if a router or another host is alive.

And sometimes a network manager needs information from another host or router.

ICMP (Internet Control Message Protocol) is a network layer protocol.

ICMP Messages are encapsulated inside IP datagrams before going to the lower layer.

Protocol field in IP header is 1 for ICMP.

22McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

ICMP Encapsulation

23McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Types of ICMP Messages

1. Error-Reporting Messages2. Query MessagesError-Reporting Messages: Chance of error always exists and

ICMP handles error reporting Error reporting messages are always

sent to the original source. ICMP uses the source IP address to

send the error message to the source (originator) of the datagram.

24McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Error-Reporting Messages

25McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Error-Reporting Messages

Destination unreachable Router can not route a datagram or a host cannot

deliver a datagram, the datagram is discard and message sent to the source.

Source Quench IP is connectionless protocol and so no control of

flow or congestion. Source has no idea whether the destination host

has been overwhelmed with datagrams. When a router or host discards a datagram due to

congestion, it sends a source-quench message to the sender of the datagram.

To inform sender about the loss of the datagram To warn the source that there is congestion in the path

and that the source should slow down the sending process.

26McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Time-exceeded: Generated in two cases When a router receives a datagram with TTL as 0, then the

router discards the datagram and sends a message to the source.

When all fragments that make up a message do not arrive at the destination host within a certain time limit.

Parameter Problem If a router or the destination host discovers an ambiguous

or missing value in any field of the datagram, it discards the the datagram and sends a message back to the source.

Redirection Routing decision on routers is made using routing protocols

but in the case of hosts, there is no routing protocol. A host may send a datagram, which is destined for another

network, to the wrong router. In this case, the router that receives the datagram will forward the datagram to the correct router. However, to update the routing table of a host, it sends a redirection message back to this host.

Error-Reporting Messages

27McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Query Messages

In the case of query message, a node sends a message that is answered in a specific format by the destination node.

Echo request and reply: for diagnostic purposes Time-stamp request and reply: to determine the round trip

time and also to synchronize the clocks in two machines. Address mask request and reply: Request by a host to

know its subnet mask to the router. Router solicitation and advertisement: Solicitation is

request by host to know the route to send a packet. Reply comes from the router as advertisement.

28McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

IPv4 Two level address structure. Inefficient. For real-time audio and video transmission,

there is no minimum delay strategies and reservation of resources.

No security mechanism [encryption and authentication of data].

IPv6 [IPng: IP next generation] Larger address space: 128 bits long. Better header format: Options are separated

from base header. This simplifies and speeds up the routing process because most of the options do not need to be checked by routers.

IPv6

29McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

IPv6 [IPng: IP next generation] New options: For additional functionalities. Allowance for extension Support for resource reservation: Type-of-

service has been removed but a mechanism called flow label has been added to enable the source to request special handling of the packet. This mechanism can be used to support traffic such as real-time audio and video.

Support for more security: encryption and authentication options in IPv6 provide confidentiality and integrity of the packet.

IPv6

30McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

IPv6 Address

Hexadecimal colon notation Divided into 8 sections, each 2 bytes in length Two bytes in hexadecimal requires four hexadecimal

digits. Abbreviation

Leading zeros of a section (four digits between two colons) can be omitted.

Only the leading zeros can be dropped, not the trailing zeros.

31McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Abbreviated Address

Abbreviations are possible if there are consecutive sections consisting of zeros only.

We can remove the zeros altogether and replace them with a double semicolon. But only once per address.

32McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

CIDR Address

IPv6 allows classless addressing and CIDR notation.

Categories of addresses Unicast: Single computer. Anycast: A group of computers with addresses that

have the same prefix. All computers connected to the same physical network share the same prefix address. A packet sent to an anycast address must be delivered to exactly one of the members of the group.

Multicast: Group of computers that may or may not share the same prefix and may or may not be connected to the same physical network. A packet sent to a multicast address must be delivered to each member of the set.

33McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Format of an IPv6 Datagram

34McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Base header: 40 bytes; Data + extension header= 65535 bytes

Ver: 4 bit; Priority: 4 bit Flow label: 24 bit; Payload length: length of IP datagram

excluding the base header. Next header: 8-bit field defining the header that follows

the base header in the datagram. Sometimes, the pointer points to the upper-layer protocol’s header.

Hop limit: TTL. Source / Destination Address: 16-bytes (128 bit). IP Add. Fragmentation in IPv6 is possible only at the source. Source must find the MTU using MTU discovery

technique. OR MTU=576 bytes (smallest possible size)

Format of an IPv6 Datagram

35McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Comparison of Network Layers in v4 and v6

ICMPv6 is designed for IPv6 Some protocols that were independent in version 4

are now part of ICMPv6. ARP and IGMP in version 4 are combined in ICMPv6. RARP is dropped from the suite because it is

seldom used.

36McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

IP Transition Strategies

Dual Stack: station should run IPv4 and IPv6 simultaneously until all the Internet uses IPv6. (Host DNS Query)

Tunneling: A strategy used when two computers using IPv6 want to communicate with each other when the packet must pass through a region that uses IPv4. To pass this region, IPv4 address is needed. IPv6 packet is encapsulated in an IPv4 packet when it enters the region, and the IPv6 packet leaves its capsule when it exits the region.

37McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Dual Stack and Tunneling

38McGraw-Hill ©The McGraw-Hill Companies, Inc., 2004

Header Translation

When the majority of the Internet has moved to IPv6 but some systems still use IPv4.

Sender wants to use IPv6, but the receiver does not understand IPv6.

Tunneling does not work in this situation because the packet must be in the IPv4 format to be understood by the receiver. In this case, the header format must be changed totally through header translation.