A vPC Type-1 inconsistency between two vPC peers in a VXLAN EVPN setup is discovered.

Configure the NVE interfaces to be Up on both switches; Configure the same VNI to multicast group mapping is the valid corrective action because the failure described is a protocol or fabric state problem, not a generic connectivity issue. For vPC, VXLAN, EVPN, NVE, VNI, Cisco troubleshooting expects the engineer to compare both endpoints of the relationship: vPC peers must agree on consistency parameters, HSRP peers must exchange hellos in the same VLAN and version context, OSPF/BGP peers must match the attributes required for stable adjacency, and VXLAN EVPN must advertise the correct host or prefix information. The selected option restores that required state. The other options either operate on the wrong layer, change a value that is not used to form the relationship, or would make the peer relationship less deterministic. A clean validation is to inspect the control-plane table first, then verify the forwarding table, and finally test end-to-end traffic. This is the Cisco-preferred approach because it avoids random changes and confirms that the fabric is forwarding based on correct protocol state.