Give metrics to see ignored/not loaded/pending-sync ingresses

[jonasbadstuebner]

I would like to have a metric/multiple metrics where I can see the state of the ingress resource from the POV of the ingress controller itself.

Background:
We had an ingress resource that was validated by the webhook, scheduled for sync by the controller, but never actually loaded into the configuration, because of the default annotation risk level allowed.
We only found out when we got an alert in our test environment, that the underlying resource could not be reached.
It took us some time to find #12635 and find out that the config option has nothing to do with the annotation validations being disabled or not.

It would have been quicker and more efficient, to get an alert for "nginx will not load this ingress as-is - check logs for more information" or something. We looked for metrics that we could build a similar alert around, but couldn't find any.

Maybe we could have a discussion, if such a metric would be useful for others too.

[k8s-ci-robot]

This issue is currently awaiting triage.

If Ingress contributors determines this is a relevant issue, they will accept it by applying the triage/accepted label and provide further guidance.

The triage/accepted label can be added by org members by writing /triage accepted in a comment.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.