WAN router or circuit fails. All devices behind create incident. Can I correlate these?
I have more than 1000 WAN routers. If a circuit or router fails I get Incidents for the switch(es) and APs behind it. I have been unable to create a rule that will create an incident for the circuit or routers while suppressing incidents for the devices that are unreachable that reside behind the failed WAN connection.
That trick not works for an every case - but if you complete the L2/L3 discovered path manually by adding the connectivity manually using nnmconnedit.ovpl, the APA will respect the path in status calculation, so nodes behind of disconnected segment are going to Unknown status. This not works for range on node types and there are special cases - if WAN router polled special way, eg only ICMP
The trick that always works for me is to manage by an interface monitoring rule the interface that connects these topologically isolated islands to the rest of your network. Just check the box in this rule after describing the criteria that matches your wan connection:
|Poll Unconnected Interfaces|
Its been working for thousands of locations for us for years. We get the topological suppression of the isolated nodes.
I've lasted longer in the technology industry than most certifications.