The health of your ``keepalived`` instances can be automatically monitored via
a bash script that verifies connectivity to all available and configured
gateway addresses. In the event that connectivity is lost, the master router
is rescheduled to another node.

If all routers lose connectivity simultaneously, the process of selecting a
new master router will be repeated in a round-robin fashion until one or more
routers have their connectivity restored.

To enable this feature, edit the ``l3_agent.ini`` file:

.. code-block:: ini

   ha_vrrp_health_check_interval = 30

Where ``ha_vrrp_health_check_interval`` indicates how often in seconds the
health check should run. The default value is ``0``, which indicates that the
check should not run at all.