ENA tracks the change in state of network objects, i.e. an up or down transition. When a network node was:
- up but is now down - ENA checks whether it is also the root cause of the failure. If it is the root cause, ENA checks whether raising a node down event is possible by associating the IP address to a device.
To identify that a node is down, all of the IP addresses on the device must be down.
When an IP address cannot be matched, e.g. the address is on an unmanaged device, ENA raises the event against a downstream device. The root cause IP address is identified in the Details field, prefixed with UPSTREAM. These events also include the list of devices impacted by the device or port failure.
- down and is now up - ENA raises the appropriate node down clear event.
By monitoring the availability of the network infrastructure over which application traffic flows, ENA can both isolate the cause of IT failures and determine their impact on application services. For example, when a router fails, this can impact on devices and applications monitored by ENA:
- when ENA manages the failed router, it raises a Network Outage event with Details identifying it as a Node Down type alarm.
- when ENA does not manage the failed router, ENA raises a Network Outage event against edge devices, with Details identifying the IP address of the failed device.
The event also identifies impacted devices and applications. Events are not generated for symptomatic alarms caused by this outage.
When identifying a device failure, ENA can also identify any monitored applications hosted by that device. In this case, rather than generating an independent alarm for each application outage, ENA raises a Network Outage event and incident against the device and indicates the applications that are impacted.