There is no such prerequisite.
infact VRF-LITE is a feature that enables a Customer to totally segregate two or more networks
and VRF-lite uses input interfaces to distinguish routes for different VPNs and forms virtual packet-forwarding tables
by associating one or more Layer 3 interfaces with each VRF.
Interfaces in a VRF can be either physical, such as Ethernet ports, or logical, such as VLAN
SVIs, but a Layer 3 interface cannot belong to more than one VRF at any time.
I hope this is useful
So i could just set up VRF on my LAN and keep everything local, very similar to VLANs. Instead of virtual networks, i would truly only have virtual routing tables.
One more question. If i do decide to set up a VRF across a WAN using OSPF, would the process ID's have to be the same on the directly connected routers?
Process IDs under OSPF are local significant whether you are using VRF or not. You don't have to match PIDs between OSPF adjacent routers.
Yes, VRF is similar to the creation of Vlans in order to segment your L2 domain. VRF segment L3 domains and routes can't be exchanged without configuring route leaking between VRFs.
Am I correct in the fact that if VRF is functioning correctly I should not be able to "ping" a VRF interface from a network not specified identified within my VRF routing table. For instance.
Router OSPF 20 VRF test
network 10.120.10.0 0.0.0.255 area 100
network 10.130.10.0 0.0.0.255 area 100
Router OSPF 40
network 172.16.50.0 0.0.0.255 area 200
If I am on a workstation within the 172.16.50.0 network, I should not be able to ping a workstation withing the 10.120.10.0 network.