The switch supports three route redistribution functions:
Intra-VRF inter-protocol route redistribution (redistribution within the same VRF instance), for example, redistribute RIP to OSPF.
Inter-VRF inter-protocol redistribution (redistribution between two VRF instances), for example, redistribute RIP in VRF 2 to OSPF in VRF 4.
Inter-VRF static routes (for example, a static route in a given VRF instance) configured as a typical static route but with the added parameter of a next-hop-vrf (the next-hop IP address is found in the next-hop-vrf instance).
With inter-VRF route redistribution, a user in one VRF instance can access route data in other VRF instances. You can redistribute routes within a VRF instance or between VRF instances; for example, one VRF instance can redistribute routes to all other VRF instances. You can redistribute Local, static, OSPF, RIP, and BGP routes and both dynamic (OSPF, BGP, and RIP) and static route redistribution is supported.
More than one routing protocol can be present in each VRF instance. Route redistribution can occur either between different protocol types, or between the same protocol types on different VRF instances.
An interface uses redistribution to announce routes that are learned by other protocols (OSPF or BGP, for example). Control route redistribution by using route policies. When you associate routing policies with route redistribution, the policy is checked before the target protocol is updated. Across VRF instances, the policy is checked at the source VRF instance, so only qualified routes are added to the routing table.
You can use static route commands to inject one specific route (including a default route) from one VRF instance to another. The route is added to the target VRF instance, while the next hop is resolved by the next-hop VRF instance.
Static routes are used to direct packets from a given source using a next-hop IP address. The next-hop-vrf option in a static route permits this path to proceed from one VRF to another. Overlapping IP addresses are supported within VRFs, thus it is possible for two VRFs to have identical IP addresses.
The following list describes interVRF route redistribution:
Redistributed routes are added to the target VRF instance, and their next hop remains in the source VRF instance.
If either the source or destination VRF instance is deleted, the redistribution configuration is automatically deleted.
Redistributed routes are not further redistributed to another VRF instance.
Route redistribution is unidirectional. You must configure route redistribution for the reverse direction if you require it. You can configure different route policies for each direction.
After you configure interVRF route redistribution between two VRF instances, you must avoid using overlapping IP addresses between these two VRF instances.
Avoid overlapping addresses; the device does not generate an error if addresses overlap.
Intra-VRF routes take precedence over inter-VRF routes.
You can physically connect two VRF instances to distribute route across VRF instances (in this case, you do not need to configure route redistribution).