Adding resourceVersion to CronJob metrics?

[JensErat]

What would you like to be added:

For deployments we already have a generation metric, which helps to see whether a deployment changed. We do not have such a metric for CronJobs -- they don't have such metadata at all. We could export the resourceVersion metadata as label though, similar to what ConfigMaps, Secrets and Ingress resources have.

Why is this needed:

#500 was closed due to being inactive, but is still an issue. We have an alert that notifies us whenever the calculated CronJob start date passed. Something like and changes(kube_cronjob_metadata_resource_version[24h]) would help to mitigate the issue.

Describe the solution you'd like

Mostly copy-paste the solution from configmaps.

Additional context

I'm fine with providing the code, if we have agreement on this being a good thing to have.

^{Jens Erat jens.erat@daimler.com, Daimler TSS GmbH, imprint}

[lilic]

👋 Thanks for the feature request! Exporting the resource version sounds good to me as we have it for other resources anyways. There is also kube_cronjob_status_active, but have not had a chance to see if that would possibly work?

cc @brancz as you commented on the last issue referenced here.