130 likes | 420 Views
Transition . There are really two types of cases that need to be addressed.Network layerHow can we get v6/v4 packets across v4/v6 networks?Host layerHow can a v6/v4 host access content on a v4/v6 host?. Network layer transition. Tunnels Dual Stack. Tunnels. Information from one protocol is enc
E N D
1. Transition and Tunnels Dale Finkelson
2. Transition There are really two types of cases that need to be addressed.
Network layer
How can we get v6/v4 packets across v4/v6 networks?
Host layer
How can a v6/v4 host access content on a v4/v6 host?
3. Network layer transition Tunnels
Dual Stack
4. Tunnels Information from one protocol is encapsulated inside the frame of another protocol.
This enables the original data to be carried over a second non-native architecture.
3 steps in creating a tunnel
Encapsulation
Decapsulation
management
5. Tunnels There are at least 4 tunnel configurations:
Router to router
Host to router
Host to host
Router to host
Required information:
V4 address of the tunnel endpoints.
Note that private addresses will not work here. Add that NATs make something breakAdd that NATs make something break
6. Tunnels How the addresses are known determines the type of tunnel.
Configured tunnel
Automatic tunnel
Multicast tunnel Take off multicast tunnel optionTake off multicast tunnel option
7. Configured tunnel These can be unidirectional or bidirectional.
Bidirectional looks like a point-to-point link
The administrator configures the tunnel.
Examples of this would be the pre-native Abilene backbone and some types of tunnel brokers.
8. Automatic Tunnel A tunnel is created without the intervention of a network administrator.
Typically this involves the v4 address of the endpoint being contained within the v6 address.
Isatap and 6to4 are examples
6to4 uses 2002::/16 plus the 32 bit v4 address to form a /48.
Isatap treats the v4 network as layer 2 transport.
V4 address is in the interface identifier -- Provide link for setting up 6to4 relay -- Provide link for setting up 6to4 relay
9. Dual Stack Obvious.
This is likely to be the predominate network layer transition tool.
When all the tools using tunnel mechanisms were developed I do not believe anyone thought viable dual stack routers would show up as quickly as they in fact have.
Most backbones will be (or could be) dual stack very easily and will be when there is a demand.
10. Transition Tunnels will remain useful as a tool for connecting isolated hosts in home networks to v6 nets. How to do this in a NAT -- toredo should fix problem, but otherwise cant be done.
Dale is going to reposition this slide as this is a problem instead of you should do this right nowHow to do this in a NAT -- toredo should fix problem, but otherwise cant be done.
Dale is going to reposition this slide as this is a problem instead of you should do this right now
11. Host level transition This is where transition could bog down.
How do you make web and other servers transparently accessable to either v6 or v4 hosts.
There are several approaches.
Dual stack
Bump-in-the-stack
Nat like devices
translators S/accessable/accessible/S/accessable/accessible/
12. Translators Within Linux variants there is a tool called Faithd.
This is a transport layer translator.
There are also header translators out there:
SIIT
Nat-PT
Socks
Various application specific translators.
S/Linux/BSD/S/Linux/BSD/
13. Summary This is neither as hard as it was once thought nor as easy as we might like to make it.
Dual Stack will be viable much sooner then was thought.
It is merely an act of faith and will to convert existing servers to v6 capable versions. If you do strange things with DNS, you may break IPv6If you do strange things with DNS, you may break IPv6