You can configure pairs of routers to act as hot standbys for each other. This redundancy is configured on an interface basis. Pairs of redundant interfaces are known as redundancy groups. The figure below depicts the active-standby device scenario. It shows how the redundancy group is configured for a pair of routers that has one outgoing interface. The Redundancy Group Configuration—Two Outgoing Interfaces figure depicts the active-active device scenario shows how two redundancy groups are configured for a pair of routers that have two outgoing interfaces.

Note that in both cases, the redundant routers are joined by a configurable control link and a data synchronization link. The control link is used to communicate the status of the routers. The data synchronization link is used to transfer stateful information from Network Address Translation (NAT) and the firewall and to synchronize the stateful database for these applications.

Also, in both cases, the pairs of redundant interfaces are configured with the same unique ID number known as the RII.

Figure 1. Redundancy Group Configuration—Two Outgoing Interfaces

Figure 2. Redundancy Group Configuration

The following scenarios are examples of Box-to-Box High Availability deployment for Cisco CSR1000v routers:

Figure 3. CSR1000v Box-to-Box High Availability on Two Independent Servers

In this deployment, two redundant Cisco CSR 1000v routers are in two independent UCS servers. The two Cisco Unified Computing System (UCS) servers can be in the same data center or two different data centers in different regions. We recommended that you configure two individual physical connections for box-to-box high availability data and control links. However, if the two dedicated physical links are not available, the box-to-box high availability data and control traffic can go through different LAN extension connections. Box-to-Box high availability parameters, such as heart beat period need to be adjusted to take into account the extended delay.

LAN interfaces of each Cisco CSR 1000v router are connected with UCS physical network interface card (NIC) interfaces through switches (for example, ESXi L2 SW). The two physical NICs on each UCS are connected to outside switch to form a box-to-box pair. Gratuitous Address Resolution Protocols (ARP) is sent from CSR LAN interfaces to reach physical switch and its Built-in Address (BIA).

Figure 4. CSR1000v Box-to-Box High Availability on Cluster Server

In the above deployment, NAT and Zone-Based Firewall (ZBFW) box-to-box high availability also works on UCS cluster setup. In this case, box-to-box control and data links go through virtual connections within the cluster. Switches (For example, ESXi L2 SW) are used to connect the 2 redundant Cisco CSR 1000v routers to form a box-to-box high availability pair; LAN interfaces on two Cisco CSR 1000v routers are connected directly to the SW switches, and two physical NICs of the cluster UCS are connected with the SW switches to communicate outside the network.

Refer to the Configuring Firewall Stateful Interchassis Redundancy module for additional information on configurations and examples.