Fortinet black logo

Administration Guide

Link monitor

Link monitor

The link monitor is a mechanism that allows the FortiGate to probe the status of a detect server in order to determine the health of the link, next hop, or the path to the server. Ping, TCP echo, UDP echo, HTTP, and TWAMP protocols can be used for the probes. Typically, the detect server is set to a stable server several hops away. Multiple servers can also be configured with options to define the protocol and weights for each server.

The link monitor serves several purposes. In the most basic configuration, it can be used to detect failures and remove routes associated with the interface and gateway to prevent traffic from routing out the failed link. More granularity is added in 7.0 that allows only the routes specified in the link monitor to be removed from the routing table. With this benefit, only traffic to specific routing destinations are removed, rather than all routing destinations.

Another enhancement starting in 7.0.1 is an option to toggle between enabling or disabling policy route updates when a link health monitor fails.

The link monitor can also monitor remote servers for HA failover. Using the HA built-in link monitor, it is only able to detect physical link failovers to trigger HA link failover. With the link monitor, remote servers can be used to monitor the health of the path to the server in order to trigger HA failover.

Finally, the link monitor can cascade the failure to other interfaces. When the update-cascade-interface option is enabled, the interface can be configured in conjunction with fail-detect enabled to trigger a link down event on other interfaces.

The following topics provide more information about the link monitor:

Link monitor

The link monitor is a mechanism that allows the FortiGate to probe the status of a detect server in order to determine the health of the link, next hop, or the path to the server. Ping, TCP echo, UDP echo, HTTP, and TWAMP protocols can be used for the probes. Typically, the detect server is set to a stable server several hops away. Multiple servers can also be configured with options to define the protocol and weights for each server.

The link monitor serves several purposes. In the most basic configuration, it can be used to detect failures and remove routes associated with the interface and gateway to prevent traffic from routing out the failed link. More granularity is added in 7.0 that allows only the routes specified in the link monitor to be removed from the routing table. With this benefit, only traffic to specific routing destinations are removed, rather than all routing destinations.

Another enhancement starting in 7.0.1 is an option to toggle between enabling or disabling policy route updates when a link health monitor fails.

The link monitor can also monitor remote servers for HA failover. Using the HA built-in link monitor, it is only able to detect physical link failovers to trigger HA link failover. With the link monitor, remote servers can be used to monitor the health of the path to the server in order to trigger HA failover.

Finally, the link monitor can cascade the failure to other interfaces. When the update-cascade-interface option is enabled, the interface can be configured in conjunction with fail-detect enabled to trigger a link down event on other interfaces.

The following topics provide more information about the link monitor: