For OMP↔BGP redistribution between two DCs with a direct interconnect, which action prevents WAN Edge routers at each DC from learning duplicate LAN routes?

Setting the same overlay AS on both DC WAN Edge routers prevents duplicate LAN routes by allowing OMP to treat the two data centers as part of the same overlay domain. When OMP sees the same AS, it understands that routes learned from one DC are internal to the overlay and avoids redistributing them back into BGP at the other DC, which would create duplicates. Defining different VRFs would segment the routing tables, but not directly prevent duplicate route learning across the interconnect. Setting the OMP admin distance lower than BGP would prioritize OMP routes, but wouldn't stop the learning of duplicates if the AS numbers differ. Setting a down-bit is used for loop prevention within the SD-WAN fabric, not for preventing duplicate route learning across a direct DC interconnect with OMP-BGP redistribution.