Link-state tracking, also known as trunk failover, is a feature that binds the link state of multiple interfaces to create redundancy in the network. The feature provides redundancy in the network when used with server network interface card (NIC) adapter teaming. When the server network adapters are configured in a primary or secondary relationship known as teaming and the link is lost on the primary interface, connectivity transparently changes to the secondary interface.
The following figure shows a network configured with link-state tracking. After you enable link-state tracking and create a link-state group, you assign interfaces to the link-state group. Interfaces connected to servers are referred to as downstream interfaces, and interfaces connected to distribution switches and network devices are referred to as upstream interfaces.
When link-state tracking is enabled, the downstream interfaces are bound to the upstream interfaces. After a set of downstream ports are associated to a set of upstream ports, if all of the upstream ports become unavailable, link-state tracking automatically puts the associated downstream ports in an error-disabled state, which causes the primary interface of the server to fail over to the secondary interface.
Figure 1. Typical Link-State Tracking Configuration
The configuration in the preceding figure ensures that when server NIC adapter teaming is used, the traffic flow continues uninterrupted when the uplink connection to a distribution switch is lost. The configuration is as follows:
The blade switches in the enclosure are connected to switch 1 and switch 2 through different switches.
Link-state group 1 is the primary link from all the blade servers in the enclosure (blade server 1 through blade server n) to switch 1.
Link-state group 2 is the secondary (backup) link from all the blade servers to switch 2.
In a link-state group, the upstream ports can become unavailable or lose connectivity because the switch or router fails, the cables are disconnected, or the link is lost. These interactions occur between the downstream and upstream interfaces when link-state tracking is enabled:
If any of the upstream interfaces are in the link-up state, the downstream interfaces can change to or remain in the link-up state.
If the server detects that the primary link is down, it redirects the traffic to the secondary (backup) link and the secondary link becomes the primary link.
As an example of a connectivity change from link-state group 1 to link-state group 2, when the primary link from blade switch 1 to switch 1 is lost, blade server 1 connects through its secondary Ethernet server interface to blade switch 2 in link-state group 2.
When link-state tracking is disabled, the entire feature is disabled. All the configuration for link-state tracking is removed. The downstream ports revert back to the state when no link-state tracking was set.
You can recover a downstream interface link-down condition by removing the failed downstream port from the link-state group. To recover for multiple downstream interfaces, disable the link-state group.