Archive for the ‘IPv6’ Category
|
Protocol |
RIPng | EIGRPv6 | OSPFv3 |
| Global Commands | ipv6 unicast-routing
ipv6 router Ripng name |
ipv6 unicast-routing
ipv6 router eigrp asn |
ipv6 unicast-routing
ipv6 router ospf process-id |
| Interface Commands | Configure IPv6 address | Configure IPv6 address | Configure IPv6 address |
| ipv6 enable | ipv6 enable | ipv6 enable | |
| ipv6 rip name enable | ipv6 eigrp asn | ipv6 ospf process-id area areanumber | |
| If router ID not automatically chosen | eigrp router-id rid
|
router-id rid | |
EIGRP for IPv6 uses the neighbor’s link local address as the next-hop and similar to RIPng in EIGRPv6 there are no classful networks so for obvious reason it cannot perform any auto summarization. A very interesting feature introduced in EIGRPv6 that EIGRP v6 does not require neighbors to be in the same subnet as a requirement to become neighbors so to form neighbor relationship In IPv6 instead of subnet mask it advertises IPv6 prefixes/length.
2002::/56 has the prefix length of 72 as (128-56) = 72 and 56 bits left for host/interface assignments.
EIGRP for IPv6 has almost same configuration procedure as for RIPng we discussed earlier.
Global Commands
ipv6 unicast-routing
ipv6 router eigrp {1 – 65535}
Interface Commands
IPv6 address xxxx.xxxx.xxxx.xxxx.xxxx.xxxx.xxxx.xxxx
Or
IPv6 address xxxx::/64 eui-64
no shut
ipv6 eigrp asn (asn is Autonomous System Number)
eigrp router-id rid
Note: In EIGRP configuration mode,.we need to configure an EIGRP router id provided there is none of IPv4 interface otherwise even if there is even one IPv4 interface then router itself create his own Router ID to use.
Selection of Router-id: The first priority would be manually assigned router-id through eigrp router-id xxxx command if it is not available then highest IP address of loopback interface and the third default priority is the highest IP address of non loopback interface also for selection all the interfaces should be in active/up state.
RIP NEXT GENERATION (RIPng)
RIP next generation is very similar to the RIP-2 version in IPv4 however there is no such concept of auto-summarization in RIPng by default in IPv6.
We can adopt following procedure to enable RIPng on Global and then on interface level.
Global Commands
ipv6 unicast-routing
It tells router that IPv6 routing is enabled on this router.
ipv6 router rip name
It defines the RIP process and (name) should be unique on that particular router atleast.
Interface level commands
IPv6 address xxxx.xxxx.xxxx.xxxx.xxxx.xxxx.xxxx.xxxx.xxxx
Or
IPv6 address xxxx.xxxx.xxxx.xxxx::/64 eui-64
either use simple IPv6 unicast addressing as in the above line or assign it through eui-64 process..
ipv6 rip name enable
Name should be same as defined earlier when defined the RIP process globally.
Verification:
To the simplest level you can verify RIPng configuration through the following commands.
Show running-config
which will show the RIPng process and if already enabled on interface.
PING IPv6 address
Will verify if IPv6 routing is being responsive.
Example:
Show running-config
interface FastEthernet0/0
no ip address (this line shows if the IPv4 address is also configured)
ipv6 address 3002::3/64
show ipv6 int brief
FastEthernet0/0 [up/up]
FE80::C000:D5FF:FED8:0 (its the link local address )
3002::3 (this is the IPv6 address)
show ipv6 int f0/0
FastEthernet0/0 is up, line protocol is up
IPv6 is enabled, link-local address is FE80::C000:D5FF:FED8:0
No Virtual link-local address(es):
Global unicast address(es):
3002::3, subnet is 3002::/64
Joined group address(es):
FF02::1
FF02::2
FF02::1:FF00:3
FF02::1:FFD8:0
MTU is 1500 bytes
Now from another connected router where the RIPng protocol is also enabled
ping ipv6 3002::3
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 3002::3, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 0/4/8 ms
Successful reply from another router will ensure the IPv6 routing and the RIPng processes between two between two routers are functional.
Purpose:
This is the simple IPv6 address configuration using GNS3 and three routers. Where most of the interfaces were assigned by using simple IPv6 address command while two interfaces in the middle router(R2) used Extended Unique Identifier (EUI) based addressing so the last 64bits were automatically assigned by the router itself after associating the MAC address of that particular interface and a step further. We used GNS3 software to demonstrate the example. Detailed configuration is also given on the image so please click on the picture for enlarged image if needed. Thank you for visiting the website.
Tips:
EUI or Extended Unique Identifier follows three simple steps to assign the interface id (the last 64 bits of the IPv6 address) which defines the 2nd part or last 64 bits of the EUI based IPv6 address of that particular interface.
Step 1. Use F0/0’s MAC Address: xxxx.xxxx.xxxx
Step 2. Split into two halves and insert FFFE: xxxx:xxFF:FExx:xxxx
Step 3. Invert 7th bit: for Hex 00 = 00000000 binary, flip for 00000010, and convert back to
hex , resulting in xyxx:xxFF:FExx:xxxx
Useful commands
For R2
show running-config
Will show how the interfaces are configured.
show interface f0/0
Will show the Hardware or MAC address of f0/0 interface
show ipv6 int f0/0
Will show three multicast groups joined by the router
FF02::1 (Address for all IPv6 devices must listen)
FF02::2 (Represents all IPv6 routers provided if IPv6 unicast is enabled. must listen)
FF02::1:FFxx:xxxx (Multipurpose Multicast address helpful for functions such as Duplicate Address Detection and Neighbor discovery; DAD and ND, must be there)
show ipv6 route
will show a new (L) LOCAL Route with 128 bit subnet mask
The Local routes, all /128 routes, are essentially host routes for the router’s unicast
IPv6 addresses. These local routes allow the router to more efficiently process packets
directed to the router itself, rather than for packets directed toward connected subnets.
debug ipv6 nd
ping 2000:0:02::3
will show NS and ND messages (Neighbor Solicitation and Neighbor Discovery Messages)
Understanding IPv6 Addressing:
Due to the fact that we are running out of 32bits IPv4 addresses in a very near future scientists suggested another efficient way to assign the IP addresses and this is IP version 6 on in short IPv6 addressing.
If all the bits are 1 in an IPv6 address then we can write it as follows
in binary format
1111111111111111.1111111111111111.1111111111111111.1111111111111111.1111111111111111.
1111111111111111.1111111111111111.1111111111111111
In IPv6 Hex format
FFFF.FFFF.FFFF.FFFF.FFFF.FFFF.FFFF.FFFF where FFFF represents one quartet or 16 bits.
Ways to write IPv6 address:
An IPv6 address FEC2.0000.0000.0000.05CD.0000.0000.04D0 can be rewritten in several ways.
FEC2::05CD.0000.0000.04D0
In IPv6 if an IP address has all consecutive 0′s in a hex quarter then we can use “::” (double column) to save some space.
Note: you can use “::” only once in an IPv6 address so it cannot be written as FEC2::05CD::04D0
FEC2::5CD.0.0.4D0
We can eliminate all the leading 0,s from the left so instead of writing fully zeroed quartet as 0000 we can use just one ‘0‘ only and 05CD and 04D0 as 5CD and 4D0.
Atlast, if the network and subnet address (prefix) is FEC2.0000.0000.0000.0000.0000.0000.0000 then to make it shorter we can write all 0′s or remaining 112 bits as follows
FEC2::/112 which is equivalent to FEC2.0000.0000.0000.0000.0000.0000.0000
Basic structure of IPv6 address:
IPv6 address consist of two parts very similar to the IPv4 address
IPv6 address = Prefix + Host id
Prefix = Network address + Subnetmask
Network address = Regionally unique bits + countrywide + ISP
For example if for Asian region assigned with first two Hex digits FE then all the IPv6 addresses in Asian region starts with FE::/120 or FE00.0000.0000.0000.0000.0000.0000.0000
Means we have 120 bits left for country wide division.
Now we suppose inside Asian region a country XYZ assigned with FEBC::/112 address where ‘BC‘ represents that particular country. Every address in that particular country should be started with FEBC.
Now we have 112 bits left for ISP’s or big governmental organizations wide division.
Lets suppose inside the country XYZ an ISP assigned with FEBC:2000::/96 so every address in that particular ISP will be started with FEBC:2000 and still we have 96 bits left for ISP wide users that is enough for 2 raise to the power 96 or 2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2x2 = 79228162514264337593543950336 number of users or 11318308930609191084 globally unique addresses for each single user anywhere in the world.These are all hypothetical calculations although very few exceptions with some specifically assigned addresses for multicast and other purposes are reserved but in general IPv6 addresses have no shortage at-least for several centuries so there is even no need to assign any private address for end users and anyone would be able to use his own globally unique IPv6 addresses even for their own sub-networks. IPv6 is not only easier to configure than IPv4 addresses but it also almost eliminated the concept of NAT inside/outside Network Address Translation as mostly the IPv6 addresses are unique all over the world. Broadcast concept is also completely eliminated and instead it uses multicast to advertise the packet to several or all users. I will discuss the IPv6 types of packets and how to assign the host portion of IPv6 addresses in the upcoming articles very soon.