4. Networking Examples : 4.6 VXLAN ORG Network for Disaster Recovery : 4.6.5 Design Considerations
4.6.5 Design Considerations
*The vCloud Networking and Security Edge (Edge) devices can be available in only one site at a time. Having the hosts in maintenance mode on the recovery side enforces this and also keeps them in sync with all of the changes that happen at the primary site.
*Removal of the primary interface after failover avoids loss of traffic for the primary site to a nonexistent interface in the case of partial environment recovery. This is because Edge always prefers the locally attached interface for sending traffic instead of sending all traffic out of the default interface. The designation of a default interface is used only for cases where there are no locally attached networks. The Edge device uses this interface to send all unknown destination traffic to its designated default router. It is prudent in the case where the environment might be failing back to its primary site to save all interface-based rules and configurations before removal. This is because all configurations associated with the interface are removed as soon as an interface is removed.
*vCloud primary sites that have direct organization networks do not use VXLAN so the recovery process is identical to the existing vCloud DR recovery process. All of the vApps on that network must have IP addresses reassigned to the correct addressing used in the recovery site Layer 3 network.
*vCloud primary sites that use isolated networks that are VLAN-backed must be recreated at the recovery site using the associated VLAN IDs available at the recovery site and the vApps reconnected to the new network. If the Isolated networks were port group-backed, the port groups still exist in the recovery site, but their definitions must be revisited to verify that their configurations remain valid.
*vCloud primary sites that use routed or isolated VXLAN-backed networks are relatively easy to recover due to VXLAN technology.