TRANSITION FROM IPV4 TO IPV6 Bachelor's thesis

TRANSITION FROM IPV4 TO IPV6  Bachelor's thesis

Bachelor's thesis

Bachelor Of Engineering

Information Technology

2015

Bhuwan Chhetri

TRANSITION FROM IPV4 TO

IPV6

BACHELOR'S THESIS | ABSTRACT

TURKU UNIVERSITY OF APPLIED SCIENCES

Information Technology | Networking

2015 | 52

Instructor: Väänänen Ossi

Bhuwan Chhetri

ABSTRACT

Most of the Internet Service Providers, and web companies are enabling IPv6 permanently for their customers and services. Since World IPv6 launch began on 6 June 2012, global IPv6 traffic has grown by 500%. If this trend continues, in less than four years, half of the Internet users will connect with IPv6.

The main purpose of the thesis is to discuss the progress of IPv6 over the depletion of IPv4 along with features including its advantages and disadvantages. It aims to discover the best solution for a transition method and factors affecting IPv6 implementation, which is solely based on the data collected from different sources. Transition techniques are presented in this thesis elaborated with configuration and challenges. The thesis concludes that using an option like dual stack is a good possible solution since the NAT (Network Address Translation) transition appears less user friendly and has been discouraged by network operators with elapse of time.

KEYWORDS:

ISPs, IPv6, IPv4, NAT, dual stack, transition

CONTENTS

LIST OF ABBREVIATIONS (OR) SYMBOLS

1 INTRODUCTION

2 INTERNET PROTOCOL

2.1

Overview

2.2

OSI Model

2.3

TCP/IP

2.4

Comparison of OSI Model and TCP/IP

3 IPV4

3.1

Features of IPv4

4 IPV6

4.1

Features of IPv6

5 COMPARISON OF IPV4 AND IPV6

6 TRANSITION FROM IPV4 TO IPV6

6.1

Dual Stack

6.2

Tunneling

6.3

6to4 Tunnels

6.4

ISTAP

6.5

Network Address Translation

6.6

Future of IP Addressing

7 ADVANTAGES AND DISADVATAGES OF PRACTICAL TRANSITION OF IPV6

OVER IPV4 47

8 CONCLUSION

26

26

28

33

35

38

45

REFERENCES

50

51

19

20

22

9

9

10

11

13

15

15

5

8

FIGURES

Figure 1. The 7 layers of OSI model [2].

Figure 2. Comparison of TCP/IP and OSI Model [6]

Figure 3. Contents of header of IPv4 packet

Figure 4. IPv4 format

Figure 5. IPv4 application using the IPv4 stack /application using both IPv4 and IPv6

10

14

15

17 stacks [10]

Figure 6. IPv6 tunneling involving different scenarios [10]

Figure 7. Tunnel consisting of protocols [10]

Figure 8. Use of IPv6 over IPv4 tunnels [10]

Figure 9. 6to4 tunnel [9]

Figure 10. ISATAP Tunnel [10]

Figure 11. IPV6-only Network Accessing IPv4 and IPv6 Internet [10]

Figure 12. Dynamic NAT-PT Operation [14]

Figure 13. Dynamic NAT-PT Operation [14]

33

36

39

42

43

26

28

29

30

TABLES

Table 1. Comparison of OSI Model and TCP/IP [5]

Table 2. Describing 5 classes of IPv4 and their area of purpose

Table 4. Comparisons of IPv4 and IPv6 [9]

Table 5. Types of IPv6 tunnels [10]

Table 6. Configuration commands for manual tunneling [10]

Table 7. Configuration commands for 6to4 tunnel [10]

Table 8. Configuration commands for an ISATAP tunnel [10]

Table 9. Basic commands used to configure the router for NAT64 [10]

Table 10. Configuration commands for static NAT-PT [10]

Table 11. Configuration commands for dynamic NAT-PT [10]

13

18

22

30

31

34

37

40

43

44

LIST OF ABBREVIATIONS (OR) SYMBOLS

IP

IPng

IPv4

IPv6

ISP

QOS

Internet Protocol

IP the next generation

IP version 4

IP version 6

Internet Service Provider

Quality of Service

HTTP

ECN

TTL

IETF

NIC

ISO

Network Interface Card

TCP Transmission Control Protocol

International Organization for Standardization

Open Systems Interconnection OSI

UDP User Datagram Protocol

Hypertext Transfer Protocol

Explicit Congestion Notification

Time To Live

Internet Engineering Task Force

OSPF

ISATAP

RFC

RARP

ARP

NAT

L2TP

MTU

PPP

RIP

DHCP

DNS

IGMP

ICMP

SNMP

VPN

Network Address Translation

Dynamic Host Configuration Protocol

Domain Name System

Internet Group Management Protocol

Internet Control Message Protocol

Layer Two Tunnel Protocol

Maximum Transmission Unit

Point-to-Point Protocol

Routing Information Protocol

Simple Network Management Protocol

Virtual Private Network

Open Shortest Path First

Intra-Site Automatic Tunnel Addressing Protocol

Request for Comments

Reverse Address Resolution Protocol

Address Resolution Protocol

ICP

CP

SP

HA

CPE

Internet Content Provider

Content Provider

Service Provider

High Availability

Customer Premise Equipment

8

1 INTRODUCTION

The IPv6 transition mechanism is a technology, which is designed to permit hosts on an IPv4 network to communicate with the hosts on an IPv6 network. With the exceptional expansion of Internet users in recent years, densely populated countries, for example China and India, are running out of IPv4 addressees. The TCP/IP has also played an important role in the global expansion of communications. More users joining the Internet results in spreading the knowledge in every field throughout the world. However, there is problem of the limited IP addresses while we are on IPv4. It has been estimated that IPv4 addressed would run out by 2011, so IPv6 was designed to solve the problem.[10]

The purpose of this thesis is to discuss the transition method and factors, which affect the IPv6 implementation. It also analyzes the progress of IPv6 and transition problems that are faced by the Internet Service Providers during the deployment of

IPv6 and how the problems can be minimized.

This thesis introduces Internet Protocol, brief introduction about the Open Systems

Interconnection (OSI) model and Transmission Control Protocol /Internet Protocol

(TCP/IP). Chapter 3 discusses the IPv4 and its feature while in chapter 4, IPv6 and features are discussed. Chapter 5 compares IPv4 and IPv6 in routing information as well virtual private network to identify its implementation differences.

Chapter 6 provides a complete transition method for the transition of IPv4 to IPv6, which are commonly used by the ISP. Section 6.6 describes the future of IP addressing in brief.

Chapter 7 analyzes the practical transition of IPv6 over IPv4.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

9

2 INTERNET PROTOCOL

2.1 Overview

On the Internet, every computer has a unique address. This address is called IP, which stands for Internet Protocol. It defines the format of packets and provides an addressing system, which has two functions: identifying hosts and providing a logical location service.

The dominant version of Internet Protocol is IPv4 and the next generation is Internet

Protocol Version 6 (IPv6). Now, one may wonder where the IPv5 is. After the IPv4,

IPv5 was introduced to overcome the obstacles or problems of IPv4. Mainly, it was designed to provide Quality of Service (Qos) for streaming services. It was envisioned to be the connection-oriented complement to IPv4 but was never introduced for public use. The next generation of Internet Protocol is IPv6, which is also called IPng or IP next generation. The features of IPv4 and IPv6 will be discussed later.

Before we go further, it is important to discuss what a network is. A network is the group of two or more computers connected with each other to exchange data using cable or wireless. When connected to the network, a computer is online and when disconnected, it is offline. In a network we can exchange different resources like data, application, hardware and other information. For this exchange, we need the following hardware: cables, switch, routers, network interface card (NIC) etc. Other than hardware, we need a server computer and a client computer. A server computers shares scanners, printers and other network services that have Internet access. Client computers are all the other computers, which are in same network with the Server computer. They can access all the resources provided by the server.

To connect to the network, the computer needs a NIC. The NIC in the computers physically connects to the network with the help of an Ethernet cable. The Ethernet cable does not connect to network but it only helps to connect to the switch while the switch connects to the network.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

10

2.2 OSI Model

The Open Systems Interconnection (OSI) model defines how the communication between two users in the network happens. The OSI model is composed of seven layers, each one having their own functions. The International Organization developed this model for Standardization (ISO) in 1984. Now, every network is built on this OSI model. The main purpose of building this model is to understand the networks and how it works. Having different layers it makes easy to troubleshoot if we have any problems in the future. The different layers of the OSI model are independent of each other but provide service to the other layers that are connected.

Again, these seven layers are divided into two parts: the four upper layers and lower three layers. The four upper layers are used when a message passes from or to a user. The three lower layers are used when a message passes through the host computer.

Figure 1. The 7 layers of OSI model [2].

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

11

There are seven layers in OSI model, which is shown in Figure 1. Each layer is described in detail below: - [2]

Physical layer: - It defines the physical equipment used for transferring the data across the network i.e. wires, computers, network cards, cables etc.

Data link layer: - This layer is used to control the signal that enter and leave the network cable. It deals with data framing and encapsulation.

Network layer: - It is concerned with the process of packet forwarding including routing through intermediate routers. It handles the addressing and routing of data based on logical addressing. Examples are apple talk DDP, IP, IPX.

Transport layer: - This layer is responsible for transferring the data from one point to point another without errors. Examples of the Transport layer are Transmission

Control Protocol (TCP), User Datagram Protocol (UDP) and Sequenced Packet

Exchange (SPX).

Session layer: - It manages, establishes, and terminates the connection between applications. It also deals with the session and connection coordination. Examples are NFS, NetBIOS names, RPC, SQL.

Presentation layer: - This layer converts data so that systems, which use different data formats, can exchange information. It is also called the syntax layer. Examples are encryption, ASCII, GIF, MIDI etc.

Application layer: - It represents the services that directly support applications such as software for file transfers, database access, emails and network games. It contains web-browser, FTP clients, email clients but does not include computer application software. Examples are WWW browsers, Telnet, HTTP, FTP, etc.

2.3 TCP/IP

TCP/IP is the network standard, which defines the Internet. The Internet Protocol (IP) standard explains how packets of information are exchanged over a set of networks.

It has a packet addressing method that lets any computer forward a packet over the network to other computer, which is closer to the packet

’s recipient. TCP checks the packets that are sent through the network and requests for transmissions if errors

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

12 are found. TCP/IP was designed to solve the problems before the OSI model was introduced. [3]

TCP/IP has a network model as the OSI network model does but they are not the same.

TCP/IP is a four-layered standard. The four layers of TCP/IP are explained below:

Layer 4. Application Layer

It is the top most layer of TCP/IP model. The application layer is where the network and its application-layer protocol reside. Using the application protocol layers, packets of information are exchanged between hosts and remote users can communicate.

The application layer includes many protocols like DNS (Domain Naming System),

HTTP protocol (which provides web document request and transfer), SMTP (which provides transfer of email message) and FTP (which provides transfer of files between two end systems). [4]

Layer 3. Transport Layer

It is the third layer of TCP/IP model, which resides in between the Transport Layer and the Application Layer. The Internet

’s transport layer transports application layer message between application endpoints.

The Transport Layer includes protocols like TCP (which provides connectionoriented services to its application) and UDP (which provides connectionless service to its application). [4]

Layer 2. Internet layer

The Internet layer is the second layer of the TCP/IP model. The Internet layer is between the Network Access layer and the Transport Layer. The Internet layer is responsible for moving data packets known as IP datagrams from one host to another. IP datagrams contain source and destination addresses, which are used to forward the datagrams between hosts and across network.

The Internet layer includes protocols like IP, ICMP, ARP, RARP and IGMP. [3]

Layer 1.Network Access Layer

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

13

The Network Access Layer is the first or lower layer of the TCP/IP model. The

Network Access Layer explains how data is sent physically through the network including individual bits.

The Network Access Layer includes protocols like FDDI, X.25, Frame Relay,

Ethernet, and Token Ring etc. [3]

2.4 Comparison of OSI Model and TCP/IP

Listed below are some of the major differences between the OSI model and the

TCP/IP model with a diagrammatic comparison.

Table 1. Comparison of OSI Model and TCP/IP [5]

OSI (Open System Interconnection) TCP/IP (Transmission Control

Protocol/Internet Protocol)

It has four layers. It has seven layers.

The transport layer guarantees delivery of The transport layer does not packets.

guarantee the delivery of packets.

OSI defines function of all the layers and TCP/IP is not flexible with other layers provides layer functioning.

as it is more based on protocols.

It has separate session and presentation It does not have separate session and layer.

presentation layer.

The Network layer provides both The Network layer provides connection-oriented and connectionless connectionless service.

service.

The OSI model is a generic, protocolTCP/IP protocols are considered to be independent standard.

standards.

Protocols do not fit well into this model.

Protocols fit well in this model.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

14

Table 1 provides a comparison between the OSI model and the TCP/IP model. It has attempted to present a practical comparison rather than a theoretical implementation. Besides this, the table also provides information on how the layers of these two models affect the performance and data communication including the frame encapsulation and defragmentation.

Figure 2. Comparison of TCP/IP and OSI Model [6]

Figure 2 provides the segments of model which could be considered as networks and Protocols layer as well as Application and Data Flow layers while there is flow of data traffic through various layers of TCI/IP and OSI model respectively. There are few similarities between these two models but the OSI model advances data communication as it increases hierarchy to application layer, resulting in improvement of performance and security.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

15

3 IPV4

IPv4 is the first version of Protocol to use publicly although it is the 4th version of

Internet Protocol. It was widely used in modern TCP/IP. It uses 32 bits of addresses and has the limit of 2^32 addresses. Due to the fast growth of Internet users, the

IPv4 addresses have been in depletion.

3.1 Features of IPv4

An IP packet is composed of a header section and a data section. It has no data checksum or any other footer after the data section. There are 14 fields in the header and of which 13 are required. The 14th field is optional. [7]

Version

Header Length

Identification

Service Type

Flags

Total Length

Fragment

Time To Live

Protocol

Source Address

Offset

Header Checksum

Destination Address

Options +Padding

Figure 3. Contents of header of IPv4 packet

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

16

Detailed information about the header of IPv4 packet is segmented in Figure 3 to give an overview of its content while the frame is transmitted with IPv4 packet. More specifically, the IPv4 packet is embedded with a source and destination address with a frame length and communication expiry duration to give delivery acknowledgement while transmitting. The contents of the header of an IPv4 packet are explained below.

Version - It is the 4-bit version as there is a value of 4.

Internet Header Length (IHL)

– The second field consisting of 4 bits is the Internet

Header Length. The minimum value is 5.

Differentiated Services Code Point (DSCP) - It is known as type of service. It is defined by TFC 2474 and used for real time data streaming, for example VoIP.

Explicit Congestion Notification (ECN) - It allows end-to-end notification of network congestion without dropping any packets. It works well when supported by an underlying network.

Total length-This is the length of whole packets which include header and data both in bytes.

Identification

– It is used to identify the original IP packet during transmission.

Flag - It has 3 bits. It helps to find out if the packets can be fragmented or not.

Fragment Offset - It is 13 bits long. It indicates the exact position of fragmenting the packets.

Time To Live (TTL)

– It avoids looping in the network and helps to limit the packets that can cross the router.

Protocol - It defines the next layer protocol.

Header Checksum - This field is 16 bits long and is used to check if the received packets are error free.

Source Address - It is the address of the sender or the source.

Destination Address- It is the address of the receiver.

Option - It is used only when the value of IHL is greater than 5. These may contain values such as security, time stamp, record route etc.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

17

IP addressing

IP address is an address which is unique and is used to identify the device in the network. It is composed of 32 binary bits. These 32 binary bits are broken into four octets, i.e., 8 bits each. Each octet is converted to a decimal and is separated by a dot. This is why we have IP addresses in dotted format, for example 82.50.69.83.

The IP address Huber has two parts: a network number and a host number. The network number determines which network the host computer is located. The host number determines the exact host computer in that network. [8]

Figure 4. IPv4 format

The structural format of IPv4 is shown in Figure 4 where different octets are segmented to describe the binary content of each octet and its composited bit capacity. Moreover, the decimal and binary composition helps to observe each octet.

It is not easy to remember an address in binary form like

01010010.00110010.01000101.01010011 for 82.50.69.83. So to make it easy, a unique name is assigned by the Domain Name System (DNS).

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

18

Classification of Address

There are several classes of IP Address as the requirement of host per network.

IPv4 is divided into 5 classes (A, B, C, D and E) and they are determined by the first octet of the IP address.

Table 2. Describing 5 classes of IPv4 and their area of purpose

Class

A

Format

N.H.H.H

Address range No

0.0.0.0

127.255.255.255

-

networks

126

of No of Area of hosts use

16,777,2

14

Large organizatio n

B

C

N.H.H.H

N.H.H.H

128.0.0.0

191.255.255.255

192.0.0.0

223.255.255.255

16,384

2,097,152

65,543

245

Medium organizatio n

Small organizatio n

D

E

N/A

N/A

224.0.0.0

N/A

239.255.255.255

240.0.0.0

N/A

N/A

N/A

Multicast groups

Experimen

255.255.255.255

N= Network number, H = host number tal

In Table 2, the different classes of IP addresses of version 4 are shown based on format and range of IP addresses. It also adds the subnet masks applied to each class along with application of the respective classes. Basically, IPv4 it has been categorized in four classes, which are detailed in above table.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

19

4 IPV6

The Internet Engineering Task Force (IETF) is in charge for defining Internet

Protocols standards. When they developed IPv4, many issues were not taken into consideration, such as address and security issues. Later on, in the early 1990s

IETF decided that to overcome the issues related to IPv4, they needed new versions of IP and IPng was created which is now known as IPv6. IPv6 offers many functions, which were not introduced in IPv4. Some of the improvements were increased address size, integrity communications, built in security. With many of these new features, IPv4 is supposed to be replaced by IPv6 in every network without any limitations. At the end of 1998, IPv6 was fully standardized. [1]

Above we have discussed the features of IPv4 and now we will be introducing the features of IPv6.

IPv6 header: -

Version Traffic Class Flow Label

Payload Length Next Header Hop Limit

Source Address

Destination Address

Version

– It specifies the Internet Protocol version 6.

Traffic Class

– It holds two values consisting of six and two. Six-bit values are used for differentiated services to classify the packets. Two-bit values are used for explicit congestion notification (ECN).

Flow Label

– The 20-bit flow label was created for giving real-time application. The flow label is used to detect spoofed packets.

Payload Length

– The 16-bit payload length contains the length of the data fields in bits.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

20

Next Header

– The 8-bit selector specifies the transport layer protocol and specifies the type of next header.

Hop Limit

– The 8-bit unassigned integer is decremented by 1, and when counter reaches 0 packet is discarded.

Source Address

–The 128-bit source address indicates the originator.

Destination Address

– The 128-bits source address indicates the recipient of the packet.

4.1 Features of IPv6

The features of IPv6 are listed below:

New Packet Format and Header

– The new IPv6 packet format assists to minimize packet header processing by routers. Operating nonessential and optional fields to extension headers that are placed after the IPv6 header attains this. Since IPv4 and

IPv6 packets are significantly different, the two protocols are not able to exchange the information.

Larger Address Space

– In comparison to IPv4, IPv6 uses four times more bits in the address. IPv4 uses 32 bits while IPv6 uses 128 bits. Due to the larger address space in the absence of NAT, there is less complexity in the network and this makes it simpler. Next, we can use a unique address for our every device in home and office.

Auto Configuration

– There are two types of auto configuration in IPv6. IPv6 supports both stateful and stateless auto configuration of every other host device.

Stateful auto-configuration

– This configuration is done manually for the installation and administration of nodes over a network. DHCPv6 works in the client server model and optionally provides IPv6 addresses and other configurations.

Stateless auto-configuration

– This configuration is suitable for individuals and small organizations. It can get automatically an IPv6 address and is not automatically registered in the DNS. The server should be manually configured.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

21

Security

– IPSec security is built in IPv6, which is difficult to add in IPv4. IPsec is mandatory in IPv6. It provides security in the network level or to the application, which are running on the IPv6 network, for example, sending/receiving data over

Internet, web server etc.

IPsec employs the Authentication Header and Encapsulating Security Payload

Header to provide security. The AH and ESP Header can be used according to our desire security. The AH and ESP header can be used in the tunnel mode and transport mode.

-

“tunnel mode”- This is the most commonly used method to encrypt traffic between secure IPSec gateways such as in between the Cisco router and PIX Firewall. It is also used to connect an end-to-end station running IPSec software. This is applied to the entire IP packet.

-

“transport mode” – This is used between end stations supporting IPSec, or between an end

–station and a gateway. This is assigned to the transport layer in the form of

IPv6 header, authentication Header or Encapsulating.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

22

5 COMPARISON OF IPV4 AND IPV6

It is important to note that IPv6 is much more than an extension of IPv4 addressing.

IPv6 offers many enchantments over IPv4, and Table 3 provides a detailed comparison of IPv4 and IPv6 along with its differences in application of real world. It also compares routing information as well virtual private network to identify its implementation differences.

Table 3. Comparisons of IPv4 and IPv6 [9]

Description

Address

IPv4 IPv6

32 bits long (4 bytes). 128 bits long (16 bytes). 64 bits

Composed of network and is for the network number and host portion, depends upon 64 bits for host. Form of the the address class. Form of address is the address is xxxx:xxxx:xxxx:xxxx:xxxx:xxxxx: nnn.nnn.nnn.nnn, where xxxx:xxxx, where x is

0<=nnn<=255,and each n is hexadecimal number. For ex:decimal digit. FE80:0000:0000:0000:0202:B3

Address types

For ex:- 192.168.255.255 FF:FE1E:8329

Spilt into 3 basic types Split into 3 basic types unicast, unicast, multicast and multicast and anycast address. broadcast address.

Configuration Need to be configured like Configuration is optional.

IP address and routes.

Domain Name System Application accept host Same support for IPv6. Support

(DNS) names and then use DNS to for IPv6 exits using AAAA (quad get IP address, using socket A) record type and reverse

API gethostbyname( ). lookup (IP-to-name). An

Domain for reverse lookup is in

–addr.arpa. application may elect to accept

IPv6 address from DNS (or not) and then use IPv6 to

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

23 communicate (or not).

A new getaddrinfo( ) API is used. Domain used for reverse llokup is ip6.arpa and not found then ip6.int is used.

Dynamic

(DHCP)

Host Used to dynamically obtain

Configuration Protocol an IP address and other configuration information.

DHCP does not support.

File Transfer Protocol Allows transferring files FTP does not support.

(FTP)

Internet across the networks.

Group Used to find the hosts which MLD does what IGMP in IPv4,

Management Protocol want traffic for particular but uses ICMPv6 by adding a

(IGMP) multicast group, few MLD-specific ICMPv6 type

Internet

Message

(ICMP)

IP header

Protocol network information. values.

Control Used to communicate Similar is used while Internet

Control Message Protocol

Version 6 (ICMPv6) some new attributes.

Variable length of 20-60 Fixed length of 40 bytes. bytes depending on IP options presents.

IP header options

IP header Type of Used by QoS and Different codes used to

Service byte differentiated services to designate a traffic class. designate an IPv6 traffic class.

Currently IPv6 does not support

TOS.

LAN connection

More options might accompany an IP header.

No IP header options. IPv6 adds optional extension headers.

Used by an IP interface to Can be used with any Ethernet get to the physical network. adapters and is also supported

Many type exits for example over virtual Ethernet between token ring and Ethernet. logical partitions.

Layer Two Tunnel Thought as virtual PPP and Does not support.

Protocol (L2TP) works over any supported

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

24

Maximum transmission

(MTU) line type.

Maximum transmission unit Has a lower boundary limit on unit of a link is the maximum MTU of 120 bytes. number of bytes that a particular link types such

(Ethernet, modem) supports. For IPv4, 576 is the typical minimum.

Tool to look at the status of Same support for IPv6. Netstat

TCP/IP connections, interface or routes.

Network address translation (NAT)

Basic firewall functions Does not support as it solves integrated into TCP/IP. the problem of shortage of

Packet filtering address.

It is the basic firewall Does not support in IPv6. functions, configured by the

Packet forwarding

System i Navigator.

The i5/OS TCP/IP can be It has limited support for IPv6. configured to forward IP packets.

Point to Point Protocol Supports dialup interface PPP does not support IPv6.

(PPP) over various modem and line types.

Private and public All IPv4 address are public, Temporary address can be address except for three address globally routed and have limited ranges which have been lifetime and generally are made private by IETF RFC indistinguishable from public

1918:10.*.*.* (10/8) , address.

172.16.0.0 through

172.31.255.255 (172.16/12)

, and

(192.168/16).

192.168.*.*

Private address domains are

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

25

Renumbering

Route commonly used by organization and cannot be routed throughout the

Internet.

It is done by manual It is important architectural reconfigurations with the element of IPv6 and is help of DHCP. automatic within /48 prefix.

Destination address is IPv6 routes are bound to forwarded to the next hop physical interface rather than an using line.

Default

*DFTROUTE route is interface. Route is associated with physical layer as the source address selection is completely different from IPv4.

Routing Information This protocol is supported Does not support.

Protocol (RIP) by the routed daemon.

Simple Network It is protocol for system Does not support.

Management Protocol management.

(SNMP)

Virtual Private Network Using IPsec it allows to Same support for IPv6.

(VPN) extend a secure private network over an existing public network.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

26

6 TRANSITION FROM IPV4 TO IPV6

There is not complete transition from IPv4 to IPv6 because IPv6 is not backward compatible. However, there are some technologies, which can convert IPv4 to IPv6.

The technologies that can convert IPv4 to IPv6 are described as below step by step.

6.1 Dual Stack

In this method, both IPV4 and IPv6 protocols are available in the same network node and that is why it can connect to remote servers with both technologies (IPv4, IPv6).

[10]

This technology ensures that only IPv4 node is upgraded. This technology is based on name lookup and application selection.

Figure 5. IPv4 application using the IPv4 stack /application using both IPv4 and IPv6 stacks [10]

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

Protocols in the world of Ethernet are implemented differently based on their IP version. Some could apply to IPv4 while others to IPv6. The diagrammatic sketch providing this information is shown in Figure 5.

27

6.1.1 Configuration [11]

Dual-Stack Router

IPv6 and IPv4

Network router# ipv6 unicast-routing interface Ethernet0 ip address 192.168.99.1

255.255.255.0 ipv6 address

2001:db8:213:1::1/64

IPv4 : 192.168.99.1

IPv6:2001:db8:213:1::1/64

6.1.2 Challenges

Dual stack can connect with both IPv4 and IPv6 technologies. Challenges faced during the deployment of dual stack method are outlined as below:

 IPv4 and IPv6 have different software requirement to run. For example, IPv4 runs with OSPFv2 and IPv6 runs with OSPFv3.

 In Dual stack Exchange, the device is configured in only one stack and most forward to dual stack devices, for example, routers and then back to the same segment using the other stacks and this results in insufficient bandwidth. To implement dual stack, IPv6 needs to be activated in all network elements and this will cost on redesign of the existing networks.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

28

6.2 Tunneling

For minimizing the transitions, all the routers on the way between the two IPv6 nodes do need to support IPv6. This method of transition is called tunneling. Primarily IPv6 packets are placed inside IPv4 packets then the packets are routed through the IPv4 routers.

One of the objections to integrating IPv6 into the current IPv4 networks is the ability to transport IPv6 packets over IPv4

–only networks. Tunneling or in IPv6 known as overlay tunnel can be used. IPv6 packets are encapsulated through the overlay tunnel in IPv4 packets for delivery across IPv4 infrastructure. The main disadvantage of tunneling is that it does not let communication between users of new protocols and old protocols without dual stack hosts.

Figure 6. IPv6 tunneling involving different scenarios [10]

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

29

Tunneling can involve any combinations of routers depending upon the end-points

(entry and exit points) of the tunnels. There are three scenarios: - host-to-host, hostto-router and router-to-router which are shown in Figure 6.

There are two types of protocols in a tunnel, namely transport protocol and passenger protocol.

Figure 7. Tunnel consisting of protocols [10]

Transport Protocol: IPv4 is the transport protocol where the tunnel is created. In the

IPv4 header, protocol 41 shows that the encapsulated data portion is an IPv6 packet.

Passenger Protocol: IPv6 is the passenger protocol. Protocols are encapsulated in the tunnel and carried over the tunnel.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

30

Figure 8. Use of IPv6 over IPv4 tunnels [10]

Table 4. Types of IPv6 tunnels [10]

Tunneling

Type

Manual

GRE

IPv4 compatible

6to4

Tunnel mode command

Tunnel source

Tunnel destination

Interface

Prefix or

Notes

Address

Ipv6ip

Ipv6ip 6to4

An IPv4 An address or address

IPv4 An IPv6 Can carry IPv6 address packets only. gre ip a reference

An to an address

IPv4 An IPv6 Can address IPv6. source96. carry

Ipv6ip auto-tunnel interface on

Not which IPv4 required. is

These are

Not required.

The

Connectionless

Network

Service configured. all point to interface

This can be multipoint address is

(CLNS), and many other generated types of the IPv4 tunneling address of types. the physical as

::tunnelpackets. Note:

IPv4 interface or compatible a loopback uses the ::/96 interface. prefix. Cisco recommends not to use this tunnel type

The IPv4 An IPv6 Site uses the destination address is address calculated in a per-packet address from the 2002::/16 prefix. Point to multipoint

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

ISATP Ipv6ip isatap

31 basic, using the IPv6 destination address.

- tunnels that can be used to connect isolated IPv6 sites.

An IPv6 Point to point prefix in multipoint modified

EUI-64 tunnels that can be used to format. connect

The IPv6 systems within address is a site. Allows generated an individual from the dual-stack host prefix and within a site to the tunnel communicate. source

IPv4 address.

Sites can use

IPv6 unicast address.

6.2.1 Configuration

IETF defined several protocols and techniques for establishing tunnels between dual-stack nodes. Primarily, all the tunneling performs the same function: transports the IPv6 packet inside and IPv4 packet between the two end points of tunnel.

In the following Table 5, there is an example and the commands for configuring the manual tunnel between two end points.

Table 5. Configuration commands for manual tunneling [10]

Command

Router(config)#interface tunnel tunnelnumber

Description

Enters interface configuration mode and states a tunnel interface and number

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

32

Router(config-if)# ipv6-address ipv6prefix/prefix-length[eui-64]

States the IPv6 network set to the interface and enables IPV6 processing on the interface.

Router(config-if)#

tunnel source

 ip-

States the source IPv4 address and the address / interface-type interface number

 number for the tunnel interface. The source IPv4 address must be reachable from the other side of the tunnel. The interface must be configured with an

IPv4 address if interface is specified.

Router(config-if)# tunnel destination ipaddress

Router(config-if)# tunnel mode ipv6ip

States the destination IPv4 address or host for the tunnel interface

States a manual IPv6 tunnel.

Note: the tunnel mode ipv6ip command states IPV6 as the passenger protocol and IPv4 as both the encapsulation and transport protocol for the manual IPv6 tunnel.

Router(config-if)#

ipv6

route ipv6prefix/prefix length tunnel tunnel-number

Configures a static route for the IPv6 prefix using the tunnel number specified in the interface tunnel command.

6.2.2 Challenges

Table 5 provides a description for different commands used for configuration for 6to4 manual tunnel. A brief outline of the challenges faced by the 6to4 manual tunnel in the network are listed below:

 It has to be manually configured.

 There will be potential issues with delay and latency through the tunnel.

 Tunnel destination point is unknown.

 Extra (additional) CPU load is needed for encapsulation and de-capsulation.

 There is no built in security. [12]

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

33

6.3 6to4 Tunnels

Manual tunnels are easy to configure but they do not scale well when a large number of tunnels is necessary. IETF has defined a mechanism called 6to4 to automatically connect to multiple IPv6 networks over one configured tunnel. 6to4 tunnels are defined in RFC 3056 [10], connection of IPv6 Domains via IPv4 Clouds. This 6to4 is point-to-multipoint connection. The difference between 6to4 and a manual tunnel is that the manual tunnel has to configure statically the other end of the tunnel, the tunnel IPv4 destination address while the IPv4 destination address is automatically derived from the IPv6 destination of the packet.

A single 6to4 tunnel can be used to connect to any number of IPv6 networks, any number of tunnel destinations. Below is the figure describing the 6to4 tunnel.

Figure 9. 6to4 tunnel [9]

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

34

6.3.1 Configuration

In the following Table 6, there is a description and the commands for configuring the

6to4 tunnel between two end points.

Table 6. Configuration commands for 6to4 tunnel [10]

Command

Router(config)# interface tunnel tunnelnumber

Description

States a tunnel interface and number and enables configuration mode.

Router(config-if)# ipv6-address ipv6prefix/prefix-length [eui-64]

States the IPv6 network set to the interface and enters IPv6 processing on the interface.

The IPv6 enable command can be used to create a link-local address and enable

IPv6 on the interface without specifying an IPv6 address.

Router(config-if)#

tunnel source

 ipaddress number

/interface-type

States the source IPv4 address or the interface source interface type and number for the tunnel interface. The source IPv4 address must be reachable from the other side of the tunnel. If an interface is defined, the interface must be configured with an IPv4 address.

Router(config-if)# tunnel mode ipv6ip States an IPv6 tunnel using a 6to4

6to4 address. Using the 6to4 technique IPv4 destination address will be determined.

Router(config-if)#

ipv6

route ipv6prefix/prefix-length tunnel tunnel-number

Configures a static route for the IPv6 to

6to4 prefix 2002::/16 to the specified tunnel interface. The tunnel number that is specified in the ipv6 route command should be same as the tunnel number in

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

35 the interface tunnel command.

6.3.2 Challenges

Table 6provides a description for different commands used for configuration for 6to4 tunnel. 6to4 tunnel can be used to connect to any number of IPv6 networks and any number of tunnel destinations. A brief outline of the challenges faced by the 6to4 tunnel in the network are listed below:

This mechanism requires a globally unique IPv4 address and the NAT tunnel endpoint is not allowed.

IPv4 packets to broadcast, multicast, and loopback address must not be sent through tunnel.

This technique is suitable for sites but not for individual hosts. [12]

6.4 ISTAP

ISATAP stands for Intra-Site Automatic Tunnel Addressing Protocol. ISTAP tunnels are sketched for transporting IPv6 packets within sites but not between the sites, where IPv6 infrastructure is not available. It is similar to other automatic tunneling mechanisms. ISATAP uses a well-defined IPv6 address format consisting of any /64 unicast IPv6 prefix and a 64-bit interface ID that contains the last 32 bits of the IPv6 address.

Unlike (?) other tunneling methods, it can remain between any two dual stack devices: host-to-host, host-to-router or router-to-host. The main feature of ISATAP is to provide dual stack hosts with access to the IPv6 network over IPv4 only network.

It is considered a quick and temporary solution to give access to IPv6 resources.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

36

Figure 10. ISATAP Tunnel [10]

6.4.1 Configuration

In Table 7, there is a description and the commands for configuring an ISATAP tunnel between two end points.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

37

Table 7. Configuration commands for an ISATAP tunnel [10]

Command

Router(config)# interface tunnel tunnelnumber

Router(config-if)# ipv6-address ipv6prefix/prefix-length [eui-64]

Router(config-if)# no ipv6 nd suppress-

ra

Description

States a tunnel interface and number and enables a configuration mode

States the IPv6 network assigned to the interface and enables IPv6 processing on the interface. Using the eui-64 option will create an IPv6 address using the

ISATAP EUI-64 format.

This commands re-enables the sending of IPv6 Router Advertisements to allow client auto configuration as sending of

IPv6 Router Advertisements is disabled by default on tunnel interface. The command no ipv6 nd ra suppress can also be entered.

Router(config-if)# tunnel source

 ipaddress number

/interface-type

States the source IPv4 address or the interface source interface type and number for the tunnel interface. If an interface is defined, the interface should be configured with and IPv4 address.

Router(config-if)# tunnel mode ipv6ip States an IPv6 tunnel using an ISATAP

isatap address.

6.4.2 Challenges

Table 7 provides a description for different commands used for configuration in an

ISATAP tunnel. These tunnels are designed for transporting IPv6 packets within sites but not between the sites. Small outlines of challenges faced by it in the network in order to check the incoming and outgoing network are mentioned below:

This technique requires more setup than other methods.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

38

It has some security issues.

It is designed for intra-site use not for the inter-site connectivity. [12]

6.5 Network Address Translation

Network Address Translation (NAT) method facilitates communication between IPv4only and IPv6-only network by translating two different IP address families. This method translates IPv6 from IPv4 and gives consistent Internet experience to the users by accessing contents over the Internet, which have IPv4 services. Another important feature of this method is that existing Internet service providers can provide IPv6 services just by using translation features and that is why they do not have to upgrade their whole system to IPv6 to provide IPv6 services to the end users. [13] The two types of translation are i) NAT64: Network Address Translation IPv6 to IPv4 ii) NAT-PT: Network Address Translation

– Protocol Translation

NAT is a common procedure in IPv4, commonly used to translate between private

(RFC 1918)[10] address and public IPv4 address space. NAT64 provides connection between IPv4-only networks and IPv6 only. It provides two major advantages over tunneling, provides a gradual and seamless migration to IPv6 and services transparently to IPv6 Internet users.

NAT64

NAT 64 is the technique or procedure for IPv4-to-IPv6 transition and IPv4-IPv6 coexistence. NAT64 is the replacement for the NAT-PT as documented in RFC

6144, Framework for IPv4/Ipv6 Translation. With DNS64, the fundamental goal of

NAT64 is to allow an IPv6-only client to initiate communications to an IPv4-only server.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

39

Figure 11. IPV6-only Network Accessing IPv4 and IPv6 Internet [10]

In Figure 11, the communication channel between IPv4 Internet and IPv6 Internet is presented, sketching the effect of protocol routing on router in Internet network.

There are three components to NAT64: -

NAT64 prefix: Prefixes /32, /40, /48, /56, /64, or /96 are used with a converted IPv4 address to transport the packet over the IPv6-only networks. A Prefix can be a network

–specific prefix (NSP) or a well-known prefix (WKP). An NSP refers to an organization and is generally a subnet from the organization’s prefix. The WKP for

NAT64 is 64:ff9b::/96. If NSP is not defined, NAT64 will use WKP to prepend the converted IPv4 address.

DNS64 server: The DNS64 server acts as a normal DNS server for IPv6 AAAA records but will try to detect an IPv4 A record when the AAAA record is not available.

If an A record is detected, DNS64 changes the IPv4 A record into an IPv6 AAAA record using the NAT64 prefix. It gives the feeling to the IPv6-only host that it can communicate with a server using IPv6.

NAT64 router: The NAT64 router advertises the NAT64 prefix into the IPv6-only network along with operating the translation between the IPv6-only and IPv4-only network.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

In Table 8, there is a description and the commands for configuring the router for

NAT64.

Table 8. Basic commands used to configure the router for NAT64 [10]

40

Command Description

Router(config)# interface type number States an interface type and number and enable the configuration mode. This interface faces only IPv6-only network and only configured with an IPv6 address.

Router(config-if)# ipv6 address ipv6address/prefix-length

States the IPv6 address and prefix length to be assigned.

Router(config-if)# nat64 enable Enables NAT64 translation on the interface.

Router(config-if)# interface type number States an interface type and number and enables configuration mode. This interface faces only IPv4-only network

Router(config-if)#

ip

address subnet-mask

address

Router(config-if)# nat64 enable ipv4and only configured with an IPv4 address.

States the IPv4 address and subnet mask to be assigned to the interface.

Enables NAT64 translation on the interface.

Router(config-if)# nat64 prefix stateful Explains the prefix and a prefix length for ipv6-prefix/prefix-length stateful NAT64:

Ipv6-prefix: this argument must be in the form documented in RFC 2373, where address is defined in hexadecimal using

16-bit values between colons.

/prefix-length: length of the IPv6 prefix.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

41

Router(config-if)# nat64 v4 pool poolname start-address-range end-addressrange

Router(config-if)# nat64 v6v4 list listname pool pool-name [overload]

Enables NAT64 IPv4 configuration. pool: configures and IPv4 address pool. pool-name: name of the IPv4 address pool. start-address-range: starting address of the address pool range. end-address-range: ending address of the address pool range.

Translates an IPv6 source address to

IPv4 source address and IPv4 destination address to an IPv6 destination address for NAT64: list: associates an IPv4 pool with the filtering mechanism, which decides when to apply an IPv6 address mapping. list name: the name of the IPv6 access list. pool: determines the NAT64 pool for dynamic mapping of addresses. pool-name: the name of the NAT64 pool. overload: enables NAT64 overload address translation(optional).

Router(config-if)#

ipv6 access-list

Explains an IPv6 ACL, and enters IPv6 access-list-name access list configuration mode. The

Router(config-if)# ipv6 permit ipv6address/ipv6-prefix-length argument access-list-name defines the name of the IPv6 ACL.

States the IPv6 address and prefix length to be translated.

NAT-PT: Network Address Translation

– Protocol Translation

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

42

NAT-PT is similar to the NAT system utilized in IPv4 that is frequently used for converting private (RFC 1918) IPv4 address to public IPv4 address and vice versa. It is used to convert IPv4 address to IPv6 address and vice versa. This method should be used only when there are no other techniques to allow IPv6-only devices to communicate with IPv4-only devices.

The common types of NAT-PT are discussed below: -

Static NAT-PT

The static NAT-PT uses static translation order to map one IPv6 address to one IPv4 address. One-to-one mapping of the IPv6 and Ipv4 address is statically configured on the NAT-PT router. It is useful when applications or servers require connection to the stable IPv4 address.

Figure 12. Dynamic NAT-PT Operation [14]

Dynamic NAT-PT

The dynamic NAT-PT translation technique needs at least one static mapping for the

IPv4 Domain Name System (DNS) server. After the IPv6 to IPv4 network is settled, reply packets travelling from IPv4 to IPv6 use the formerly settled dynamic mapping to convert back from IPv4 to IPv6 and vice versa for an IPv4-only host.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

43

Figure 13. Dynamic NAT-PT Operation [14]

6.5.1 Configuration

InTable 9, there is a description and the commands for configuring static NAT-PT.

Table 9. Configuration commands for static NAT-PT [10]

Commands Description

Router(config)# interface type number States an interface type and number and enable the configuration mode. This interface faces only IPv6-only network and only configured with an IPv6

Router(config-if)# ipv6 address ipv6address/prefix-length

Router(config-if)# ipv6 nat address.

States the IPv6 address and prefix length to be referring to the interface.

Enables NAT-PT on the interface.

Router(config-if)# interface type number States an interface type and number and enable the configuration mode. This interface faces only IPv4-only network and only configured with an IPv4 address.

Router(config-if)#

ip

address ipv4address subnet-mask

Router(config-if)# ipv6 nat

States the IPv4 address and subnet mask to be referring to the interface.

Enables NAT-PT on the interface.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

44

Router(config-if)# ipv6 nat prefix ipv6prefix/prefix-length

Explains the IPv6 prefix to be used as the NAT-PT prefix for the IPv6 domain.

The only prefix length supported is 96.

Router(config-if)# ipv6 nat v4v6 source Packets with the source ipv4-address will ipv4-address ipv6-address be converted to the source ipv6-address.

The IPv6-address is the destination IPv6 address used by the IPv6 host to reach the IPv4 host.

Router(config-if)# ipv6 nat v6v4 source Packets with the source IPv6-address ipv6-address ipv4-address will be converted to the source IPv4address. The IPv4-address is the destination IPv4 address used by the

IPv4 host to reach the IPv6 host.

In Table 10, there are the configuration commands for dynamic NAT-PT with the description.

Table 10. Configuration commands for dynamic NAT-PT [10]

Commands

Router(config)# ipv6 access-list ipv6acl-name

Description

Explains the IPv6 access list and enables the configuration mode in the router.

Explains the range of IPv6 addresses allowed to be translated.

Router(config-ipv6-acl)# permit sourceipv6-address/prefix-length destinationipv6-address/prefix-length)

Router(config)# ipv6 natv6v4 source list Explains the dynamic mapping between ipv6-acl-name pool ipv4-pool-name the permitted IPv6 source address and the pool of IPv4 address. The IPv6-aclname identifies the IPv6 ACL used to determine which IPv6 will be translated.

The IPv4-pool-name selects the pool of

IPv4 address to be used in the

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

45

Router(config)# ipv6 nat v6v4 pool ipv4pool-name ipv6-acl-name start-ipv4address end-ipv4-address translation.

Explains the pool of source IPv4 address used for translation.

6.5.2 Challenges

Table 10 provides a description for different commands used for configuration in static and dynamic NAT-PT. Mostly, it modifies restriction based on the updated configuration of network. It works with IPv6 access-list and allows/denies the network accordingly. Small outlines of challenges faced by NAT in the network in order to check the incoming and outgoing network are mentioned below:

 They are not expected to be used widely as they slow down packet flow.

 They do not allow the network to exploit specific capabilities of either protocol.

They act as a redundant channel in the online communication over the

Internet. [12]

6.6 Future of IP Addressing

Everyone one in the modern business world uses Internet and its rapid growth is the main reason behind the importance of Internet Protocol (IP) address. Every device must have a unique IP address to connect to the Internet in order to exchange data with another. Everyone uses IP addresses but only few understand them as they cannot be seen and are highly intangible. There is a great change happening as people are moving to adopt IPv6 and providers are trying to save their customers by moving to IPv6. If providers do not do so, they will soon be out of business as the

IPv4 will be no longer in use. Providers should understand IPv6 and ensure they are prepared for the future, IPv4 and IPv6 are not compatible and devices using one or the other cannot communicate directly. Either providers should upgrade their devices or change to IPv6 by using transition techniques. One of the popular techniques is dual stacking as it supports both IPv4 and IPv6. IPv6 is the only long-term solution to

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

46 the depletion of IPv4 address pool. The depletion IPv4 address pool affects routing security. While distributing IPv4 addresses there was always pressure to preserve the free pool of IPv4 by making smaller allotment and aggregation. The astounding amount of available IPv6 address means that conservation is no longer pressing and large piles of blocks can be given out. [15]

Without IPv6 there will be limitation in use of technology in the coming years such as smartphones, which are gaining popularity in the recent years. With the increase of population and users of Internet, it is hard to settle with IPv4. Every Internet provider and company is moving to IPv6. Most of the countries, like China, the USA and India etc., have already moved their services to IPv6. For the continuous growth of

Internet and spreading through the world, deployment of IPv6 is necessary.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

7 ADVANTAGES AND DISADVATAGES OF PRACTICAL

TRANSITION OF IPV6 OVER IPV4

47

ISPs and big companies need to take steps to ensure service continuity with transparency for their customers at all the times during transition and co-existence. It is most important that the transition to IPv6 is stable and non-interruptive to exiting services. The operators should have clear ideas how they will transition to IPv6 and know the risks and challenges ahead before they start the transition. Some operators may focus on green design while other may focus on IPv4 and IPv6 co-existence.

Operators should have list of strategies and concepts what they are seeking for.

IETF has been developing tools for more than a decade for transitioning to IPv6 but many operators have not yet begun yet. The reason behind that could be lacking

IPv6 development in applications, hosts, CPEs, network equipment, and contents.

Another reason could be lack of knowledge in applying technologies and techniques in the network without causing interruption in service. The transition problem faced by the providers include the following areas: Network, Connectivity, Applications, and network Management and Operation. [16] a) Network problems

Address Architecture

IPv6 has much larger address space in comparison with IPv4. Due to the large IPv6 address space, special attention is needed when designing the IPv6 network since it differs from the fragmented and smaller IPv4 address design. Providers will need to plan in advance for IPv6 unlike IPv4, will provide them with an enormous address space, which needs cautious architectural consideration.

Connectivity

While starting transition to IPv6, the systems engineers should design a network to provide continuity of service to the customers. For this, dual stack is the natural approach but due to the depletion of IPv4 address and cost, providers may consider to upgrade part of their network to IPv6-only.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

48

High Availability

High Availability (HA) is the major requirement for every service and network service.

Providers have huge experience in running high availability in IPv4 using mature protocols, such as VRRP and OSPF Graceful Restart. Compared to IPv4, HA for

IPv6 is less known. An application running on IPv6 may need to failover to IPv4 network due to network failure during transitioning. Some work needs to be done in this area to fix this problem. Besides, the new transition techniques require new HA models. If HA is supported, providers will normally deploy a transition method. [16] b) Application Problems

During the transition process, IPv4 and IPv6 application will coexist in the network.

Regardless of what technology providers choose to use, services should be provided to the customers. Providers should find out the best techniques for the transition without affecting the services they provide. c) Network Management and Operation Problems

In paper, organizing an IPv6 network should be similar to organizing an IPv4 network. For example, SNMP works over IPv6 without modification. New technologies and techniques may be introduced during the transition process. These new technologies and techniques require new operation models.

The growth of Internet users has led to the shortage of usable IPv4 in the nearer future. Giant communication companies, like China Telecom, are practical examples for encouragement of IPv6, because of consequences caused due to the limited

IPv4 addresses. Few technical issues arise due to the change of platform from IPv4 to IPv6 and these are outlined as follows: -

1. Limited availability of IPv4 addresses to China Telecom gave rise to the need to find a new way of identifying Internet gateways. This led to the transition to IPv6. [17]

2. Newly used IPv6 is uncommon in Internet websites and very few

Internet Content Provider (ICP) look for the option of IPv6 when

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

49 expanding proprietary services because of the enormous code change and increasing manpower. Many business websites are always linking each other and creating a complex structure, which leads to many problems when one website migrates to IPv6 only. Content

Provider/Service Providers (CP/SP) do not realize how urgent it is to migrate to IPv6, which is the main reason why ICP migration lacking motivation. [17]

3. Some specific terminals (for example, set top box) do not support IPv6 while the main operating systems do. [17]

4. China Telecom has two key problems while IPv6 transition, large-scale network and large number of subscribers. The transition involves multiples level and broad scope, which lead to huge costs in modification as with the large-scale network and various service platforms. [17]

5. The use of IPv6 becomes a reliable solution for companies like China

Telecom and Internet-based companies for expanding the volume of subscribers.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

50

8 CONCLUSION

The thesis discusses the approach of IPv6 over the limited IPv4 in Internet world where users have increased rapidly. It also compares the consequences and features of transition from IPv4 to IPv6. Transition methods with their configuration and challenges that come forward during each transition process, are also documented in this thesis. A practical approach of various transition methods led to the conclusion that dual stack remains more popular and practical with low cost in implementation and supported by wide range of devices. Transition methods, like tunneling and translation, are not optimally supported for the networks during a transition from IPv4 to IPv6 although these tools are provided by IETF to make the transition easier. Thus, dual stack seems the preferable method to begin adopting

IPv6 with upgradable devices in order to securely manage the exiting IPv4

Infrastructure. This transition incurs minimal impact on customers, as they do not have to move IPv6 overnight and can deploy and migrate to IPv6 when they are ready.

The thesis also provides the comparison between IPv4 with IPv6, reflecting on the consequences arising due to transition from one to the other. IPv6 has newly introduced an Internet algorithm, taking a comparable elapse of time to complete the upgrade on the entire telecommunication platform, and working with zero stress under IPv4. The transition problems faced by the Internet service providers concern network, connectivity, applications, and network management and operation. Before the process of deployment, a study can be carried out so that the time and cost can be saved during the transition process. The deployment of IPv6 is the best way for the growth of Internet and users around so that there will be no any limitation to using the technology that are discovering.

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

51

REFERENCES

[1] What is IPv6? [ONLINE] Available at: http://IPv4.opus1.com/IPV6/whatisIPV6.html

.

[Accessed: 03 March 2015].

[2] Router-Switch. 2013.

What is OSI Model & the Overall Explanation of ISO 7 Layers.

[ONLINE] Available at: http://IPv4.cisco1900router.com/what-is-ios-model-the-overallexplanation-of-ios-7-layers.html

. [Accessed: 04 March 2015].

[3] omnisecu. TCP/IP. [ONLINE] Available at: http://www.omnisecu.com/tcpip/tcpip-model.php

.

[Accessed: 07 July 2015].

[4] Kurose, J &Ross, K. 2010. Computer Networking. 5th ed. USA: Pearson

[5]

Eazynotes. Comparasion between OSI and TCP/IP model. [ONLINE] Available at: http://www.eazynotes.com/notes/computer-networks/slides/comparison-between-osi-tcpipmodel.pdf

. [Accessed: 04 July 2015].

[6] tutorialslodge. 2014. OSI and TCP/IP Model. at: http://tutorialslodge.com/osi-tcpip-model/ . [Accessed: 10 July 2015].

[ONLINE] Available

[7] tcpipguide.

Internet Protocol Version 4 (IP, IPv4). [ONLINE] Available at: http://IPv4.tcpipguide.com/free/t_InternetProtocolVersion4IPIPV4.htm

. [Accessed: 20 May

2015].

[8] IPv4 Addressing. 2015.

IPv4

Addressing. [ONLINE] Available at: https://technet.microsoft.com/en-us/library/dd379547(v=ws.10).aspx

. [Accessed: 02 July 2015].

[9] IBM.

Comparison of IPv4 and IPv6. [ONLINE] Available at: https://IPv4-

01.ibm.com/support/knowledgecenter/ssw_ibm_i_72/rzai2/rzai2compIPV4IPV6.htm

. [Accessed:

08 April 2015].

[10] Graziani. R. 2012.

IPv6 Fundamentals: A Straightforward Approach to Understanding IPv6.

1 Edition. Cisco Press.

[11] training.apnic.net. 1800.

IPv4 to IPv6 Transition. [ONLINE] Available at: https://training.apnic.net/docs/eIP603_Transition.pdf

. [Accessed: 12 July 2015].

[12] Sellers, C. 2009.

IPv6 Transition Mechanisms and Strategies. [ONLINE] Available at: http://IPv4.rmv6tf.org/wp-content/uploads/2012/11/Chuck-Sellers-090421-IPV6-Transition-

Mechanisms-Sellers1.pdf

. [Accessed: 29 June 2015].

[13] Das, K. Network Address Translation (NAT) Pros & Cons. Network Address Translation

(NAT) Pros & Cons, [Online]. Available at: http://IPv6.com/articles/nat/NAT-Pros-and-Cons.html

[Accessed: 29 July 2015].

[14] Cisco. NAT-PT for IPv6. [ONLINE] Available at: http://IPv4.cisco.com/c/en/us/td/docs/iosxml/ios/ipaddr_nat/configuration/15-mt/nat-15-mt-book/ip6-natpt.html

. [Accessed 28 July 2015].

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

52

[15]

IP Addresses. 2013. IP Addresses. [ONLINE] Available at: http://www.itproportal.com/2013/07/02/ip-addresses-how-they-shaped-the-past-of-theinternet-and-what-they-will-influence-its-future/ . [Accessed: 09 August 2015].

[16]

IETF. 2010. Problem Statements of IPv6 Transition of ISP. [ONLINE] Available at: https://tools.ietf.org/html/draft-lee-v4v6tran-problem-02 . [Accessed: 04 August 2015].

[17] Internet Engineering Task Force. 2011. Broadband Service Provider Use Case

. [ONLINE]

Available at: http://www.watersprings.org/pub/id/draft-tian-v4v6tran-broadband-sp-usecase-

00.txt

. [Accessed 11 July 2015].

TURKU UNIVERSITY OF APPLIED SCIENCES THESIS | Bhuwan Chhetri

Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement