After adding the verbosity option to the cluster autoscaler (in OCPCLOUD-1571), the default verbosity was changed from 3 to 1 (there is additional info in KCS 6172162 [1]).

The current log verbosity is not showing almost any useful information about why the cluster is not scaling up/scaling down, and KCSs like [2] (and the possible insights rules) will not work unless customer increases the verbosity.

4.12

Always

The current verbosity is not showing useful information about the Cluster Autoscaler.

Show useful information about the Cluster Autoscaler like in previous releases.

Currently, the Cluster Autoscaler is too verbose, but it's not showing useful information.
For example, the message `Scale-down calculation: ignoring 4 nodes unremovable in the last 5m0s` is shown every 10 seconds, but it's not providing any useful information:

~~~
I0101 01:11:38.460152       1 static_autoscaler.go:419] No unschedulable pods
I0101 01:11:38.865989       1 scale_down.go:509] Scale-down calculation: ignoring 4 nodes unremovable in the last 5m0s
I0101 01:11:39.274789       1 scale_down.go:918] No candidates for scale down
I0101 01:11:50.298303       1 static_autoscaler.go:419] No unschedulable pods
I0101 01:11:50.700662       1 scale_down.go:509] Scale-down calculation: ignoring 4 nodes unremovable in the last 5m0s
I0101 01:11:51.100425       1 scale_down.go:918] No candidates for scale down
I0101 01:12:02.116608       1 static_autoscaler.go:419] No unschedulable pods
I0101 01:12:02.516659       1 scale_down.go:509] Scale-down calculation: ignoring 4 nodes unremovable in the last 5m0s
I0101 01:12:02.913941       1 scale_down.go:918] No candidates for scale down
I0101 01:12:13.927574       1 static_autoscaler.go:419] No unschedulable pods
I0101 01:12:14.330182       1 scale_down.go:509] Scale-down calculation: ignoring 4 nodes unremovable in the last 5m0s
I0101 01:12:14.726754       1 scale_down.go:918] No candidates for scale down
I0101 01:12:25.765062       1 static_autoscaler.go:419] No unschedulable pods
I0101 01:12:26.169995       1 scale_down.go:509] Scale-down calculation: ignoring 3 nodes unremovable in the last 5m0s
I0101 01:12:26.170199       1 scale_down.go:613] 1 nodes found to be unremovable in simulation, will re-check them at 2023-01-01 01:17:24.948558791 +0000 UTC m=+18438792.813544309
I0101 01:12:26.556117       1 scale_down.go:918] No candidates for scale down
I0101 01:12:37.571027       1 static_autoscaler.go:419] No unschedulable pods
I0101 01:12:37.973343       1 scale_down.go:509] Scale-down calculation: ignoring 4 nodes unremovable in the last 5m0s
I0101 01:12:38.371852       1 scale_down.go:918] No candidates for scale down
I0101 01:12:49.386368       1 static_autoscaler.go:419] No unschedulable pods
I0101 01:12:49.789081       1 scale_down.go:509] Scale-down calculation: ignoring 4 nodes unremovable in the last 5m0s
I0101 01:12:50.185197       1 scale_down.go:918] No candidates for scale down
I0101 01:13:01.213753       1 static_autoscaler.go:419] No unschedulable pods
I0101 01:13:01.613837       1 scale_down.go:509] Scale-down calculation: ignoring 4 nodes unremovable in the last 5m0s
I0101 01:13:02.012856       1 scale_down.go:918] No candidates for scale down
~~~

Can those messages be reduced to each 60 seconds instead, and show more useful messages of the reason for not scaling up/scaling down?