When operating an MC1500, Ethernet redundancy can be enabled at any time by simply following the steps outlined in the following sections. However, for the following controller models enable dual port bonding before activating Ethernet redundancy:
- MC5000 (with accelerator card)
To enable dual bonding, enter the following commands and reboot the controller:
default# configure terminal default(config)# bonding dual default(config)# exit default# copy running‐config startup‐config
Configure Redundant Ethernet Failover With the CLI
The following commands configure Ethernet interface 2 on a controller as a backup to Ethernet interface 1:
default# configure terminal default(config)# interface FastEthernet 2 default(config‐if‐FastEth)# type redundant default(config‐if‐FastEth)# exit default(config)# exit
default# copy running‐config startup‐config
In the redundant configuration, the IP address for the second Ethernet interface cannot be configured. It will receive the IP address of the primary Ethernet interface when the failover occurs.
The system requires a reboot for the change to become effective. Reboot the system now, and then check the redundant second interface configuration with the show second_interface_status command: default# show second_interface_status
Recovering From Redundant Ethernet Failover
Once Dual Ethernet Redundant mode configuration is complete, the controller needs to be rebooted – see directions above. After the reboot, if the first Ethernet interface link goes down, then the second Ethernet interface takes over the controller connectivity. Redundant Ethernet failover is based on LinkID and does not require any spanning-tree configuration. When a LinkID is missing, the failover will occur in under one second. This failover will be transparent to the access points. The second interface remains active and serving all APs, even if the first interface comes up again. Verify this with the CLI command show second-interface-status. Only when the second interface goes down will the first interface (if it is up) take over the controller connectivity.
In hardware controllers bringing the switch port down will be detected as interface down and a link down alarm will be generated, rather in a virtual controller bringing the switch port down will not be detected as interface down and hence no link down alarm will be generated.
An alarm will be generated when the mapped interface in the VMWare client software is configured as disconnected.
When N+1 or L3 redundancy is also configured and controller 1 fails, the APs move to controller 2. When controller 1 comes back online, the APs immediately begin to move back to controller 2. Also see Recovering From N+1 with Dual Ethernet Failover.
Having trouble configuring your Fortinet hardware or have some questions you need answered? Ask your questions in the comments below!!! Want someone else to deal with it for you? Get some consulting from Fortinet GURU!