CSIS 4823 Data Communications Networking IPv 6 Mr

  • Slides: 24
Download presentation
CSIS 4823 Data Communications Networking – IPv 6 Mr. Mark Welton

CSIS 4823 Data Communications Networking – IPv 6 Mr. Mark Welton

IPv 6 � Why IPv 6? � Ipv 4 is a 32 -bit addresses

IPv 6 � Why IPv 6? � Ipv 4 is a 32 -bit addresses � This is 4, 294, 967, 296 (4. 2 x 109) IP addresses � Large blocks are considered wither reserved or unroutable (i. e. 10. 0/8 private address space) � There are currently 7 billion people on the Earth � We could not even have every person get a device with a public IP under IPv 4

IPv 6 � IPv 6 established in 1998 by RFC 2460 � Changes from

IPv 6 � IPv 6 established in 1998 by RFC 2460 � Changes from IPv 4 � Security – extensions for authentication, data integrity, and confidentiality � Flow labels – classifying packets belonging to traffic flows � More options – more room for additional IP header options and extensions � Header format – reduced the required fields in the IP header � IP address space – 128 bits and “anycast” address, multicasts, and broadcast

IPv 6 addressing � IPv 6 uses 128 -bit addresses � This equates to

IPv 6 addressing � IPv 6 uses 128 -bit addresses � This equates to 3. 40292367 x 1038 addresses � Roughly 295 IP addresses for every person on the Earth � Address scheme was designed to support efficient route aggregation � Subnets in IPv 6 are suppose to contain 264 addresses � This would mean that the IPv 4 space would consume 1/64 th of the address space available in just one IPv 6 subnet

IPv 6 addressing � Addresses do not use dotted decimal notation, but rather hexadecimal

IPv 6 addressing � Addresses do not use dotted decimal notation, but rather hexadecimal � IPv 6 address is commonly shown in four digit hexadecimal blocks � Each hexadecimal block is 16 -bits � AAAA: BBBB: CCCC: 0000: 11111: 1 111

IPv 6 addressing � IPv 6 shorthand allows for leading zeroes within a block

IPv 6 addressing � IPv 6 shorthand allows for leading zeroes within a block can be eliminated � AAAA: BBBB: CCCC: 0000: 11111: 1 111 � So this address would become � AAAA: BBBB: CCCC: 0: 0: 11111: 1111

IPv 6 addressing � IPv 6 shorthand also allows any consecutive number of zeros

IPv 6 addressing � IPv 6 shorthand also allows any consecutive number of zeros to be replaced by double colon � so � AAAA: BBBB: CCCC: 0: 0: 11111: 1111 � would be shown as � AAAA: BBBB: CCCC: : 1111: 1111

IPv 6 addressing � the following three addresses are identical: � AA 76: 0000:

IPv 6 addressing � the following three addresses are identical: � AA 76: 0000: 0012: A 322: FE 33: 2267 � AA 76: 0: 0: 0: 12: A 322: FE 33: 2267 � AA 76: : 12: A 322: FE 33: 2267

IPv 6 addressing � Why hexadecimal? � HEX: AA 76: 0000: 0012: A 322:

IPv 6 addressing � Why hexadecimal? � HEX: AA 76: 0000: 0012: A 322: FE 33: 2267 � If we represent this IPv 6 address in decimal it would have 12 more octets them a IPv 4 address � DEC: 170. 118. 0. 0. 18. 163. 34. 254. 51. 34. 103

IPv 6 addressing � IPv 6 is not backwards compatible to IPv 4 (you

IPv 6 addressing � IPv 6 is not backwards compatible to IPv 4 (you can not represent an IPv 6 address in IPv 4) � IPv 4 addresses are compatible with IPv 6 addresses ◦ This is done by setting the first 80 bits to 0 and the next 16 bits to 1, with the final bits being the IPv 4 address � the IPv 4 address of 10. 20. 30. 40 would be: � 0000: 0000: ffff: 10. 20. 30. 40 � 0: 0: 0: ffff: 10. 20. 30. 40 � : : ffff: 10. 20. 30. 40

IPv 6 Subnet Masks � Masking IPv 6 addresses is similar to IPv 4

IPv 6 Subnet Masks � Masking IPv 6 addresses is similar to IPv 4 in that we use prefix lengths � To understand it first you must know the rules for IPv 6 addressing

IPv 6 Subnet Masks � The first 48 bits of IPv 6 addresses are

IPv 6 Subnet Masks � The first 48 bits of IPv 6 addresses are reserved for Internet routing (network address range) ◦ ffff: 0000: 0000 � The 16 bits from bit number 49 to bit number 54 are for defining subnets (subnetting range) ◦ 0000: ffff: 0000: 0000 � The last 64 bits are for device addresses ◦ 0000: ffff: ffff

IPv 6 Subnet Masks The most common subnet mask is and should be /64

IPv 6 Subnet Masks The most common subnet mask is and should be /64 � The American Registry for Internet Numbers has the following guidelines for assigning IPv 6 address space: � /64 – when it is known that only one subnet is needed � /56 – for small sites needing a few subnet within five years � /48 - for large sites � � � IETF (Internet Engineering Task Force) recommends all IPv 6 networks be /64 IPv 6 does not support NAT

IPv 6 Address Types � Like IPv 4, certain addresses are special and are

IPv 6 Address Types � Like IPv 4, certain addresses are special and are identified by their high-order bits � These are defined in RFC 4291 Address type Binary prefix IPv 6 notation Section Unspecified 00. . . 0 (128 bits) : : /128 2. 5. 2 Loopback 00. . . 1 (128 bits) : : 1/128 2. 5. 3 Multicast 1111 FF 00: : /8 2. 7 Link-local unicast 1111111010 FE 80: : /10 2. 5. 6 Global unicast (everything else)

IPv 6 Address Types � Note the /128 this is the IPv 6 version

IPv 6 Address Types � Note the /128 this is the IPv 6 version of a /32 in IPv 4 � This specifies a host address (mask of all ones) � A /32 means a large aggregate Internet route in IPv 6 not a host address

IPv 6 Address Types � Unspecified � An address unspecified address is an address

IPv 6 Address Types � Unspecified � An address unspecified address is an address of all zeros � Packets with an unspecified destination will be dropped by routers � The source address may be unspecified by devices that have yet to learn their address

IPv 6 Address Types � Loopback � The address loopback address in IPv 6

IPv 6 Address Types � Loopback � The address loopback address in IPv 6 is 0: 0: 1 or : : 1/128 � This is analogous to 127. 0. 0. 1 in IPv 4

IPv 6 Address Types � Multicast is a bit different in IPv 6, though

IPv 6 Address Types � Multicast is a bit different in IPv 6, though the basic principle is the same � The first thing you should notice is that there is no concept of a broadcast in IPv 6 � If you want to send a packet to all nodes, you should use the link-local all nodes multicast address of ff 02: : 1

IPv 6 Address Types � Global unicast addresses are what you might call “normal”

IPv 6 Address Types � Global unicast addresses are what you might call “normal” addresses, and will be assigned to interfaces on your devices

IPv 6 Address Types � Link-local unicast addresses are described in RFC 2373 (section

IPv 6 Address Types � Link-local unicast addresses are described in RFC 2373 (section 2. 5. 8) as follows: ◦ Link-Local addresses are designed to be used for addressing on a single link for purposes such as automatic address configuration, neighbor discovery, or when no routers are present. Routers must not forward any packets with Link-Local source or destination addresses to other links. � So what does that all mean?

IPv 6 Address Types � Link-local � It unicast is the IPv 6 equivalent

IPv 6 Address Types � Link-local � It unicast is the IPv 6 equivalent of the 169. 254. x. x address space � In IPv 6 this will be in the fe 80: : /10 network � How do we create a unique address?

IPv 6 Address Types � Link-local � We unicast use the routing prefix combine

IPv 6 Address Types � Link-local � We unicast use the routing prefix combine with the MAC address � How many bits are in a MAC address? � How many bit are in the device part of the address?

IPv 6 Address Types � Link-local � We unicast pad the 48 bit MAC

IPv 6 Address Types � Link-local � We unicast pad the 48 bit MAC address with ff: fe in the middle to create the Extended Unique Identifier 64 (EUI-64) � We must also flip the seventh bit in the first octet of the MAC address � This bit is the locally administered bit used to indicate that the OS has altered the address

IPv 6 Address Types

IPv 6 Address Types