This might be a weird case, but I use icinga2 to monitor my home network, but I also like to periodically suspend my machines.
When a host is down because it’s suspended, it shows up properly in the ‘host problems’ list, where I can glance and ignore.
When the host is down, though, I have to sometimes take down dependency services (nfs mounts, etc). This happens automatically as the service’s host detects that the mount is unavailable.
I have this service dependency that seems to be working:
apply Dependency "nfs-depdnancy" to Service {
parent_host_name = "tesla.lan"
parent_service_name = "nfs-server"
states = [ OK, Warning ]
disable_checks = true
disable_notifications = true
assign where service.vars.nfs_dependancy
}
Now I’ve also noticed that when a host is down, the services under that host are marked ‘handled’, but this doesn’t seem to happen with service dependencies.
Is there not a way to have service dependencies handled when the host is down? Then I can easily ignore them, since the real issue is the host, not the service.
I should say, I’ve kinda been using the interface for notifications more than the email, so the visual effect of a handled service is convenient to me, vs a critical service that isn’t handled.