I've noticed a strange thing with the Image Registry running on Azure in OpenShift 4.2.7 (possibly all other versions too).
When the registry is idle, I'm seeing about 40 requests per minute for "List Storage Account Keys" per minute in Azure console, under the resource group "Activity log". Each request is has a "Started" and "Succeeded" event, so it's 80 events in total per minute.
This to me seems hugely excessive and unnecessary, is it expected behaviour?
Surely listing storage account keys only needs to happen a few times per day instead of almost per second?
I looked in the logs of the "cluster-image-registry-operator" and these log entries seem to correlate within about a second of the azure activity log entries.
I1125 05:20:33.804388 14 controller.go:260] event from workqueue successfully processed
I1125 05:20:36.816148 14 controller.go:260] event from workqueue successfully processed
I1125 05:20:39.987651 14 controller.go:260] event from workqueue successfully processed
Should I raise a bugzilla for this?