Allow scaling of workload clusters if there are critical updates available.

[Elektronenvolt]

Title: Allow scaling of workload cluster nodepools if there are critical updates available.

Description: I wanted to scale up nodepools of a workload cluster with Set-AksHciNodePool -clusterName <cluster> -name <nodepool> -count <n> and got stopped by an error message to do a critical upgrade first.

It would be nice to allow nodepool scaling if there are any critical upgrades available but not installed yet. In case of running out of resources, being forced to do a cluster upgrade first can cause serious trouble. In situations where peak loads are coming in not being able to scale up nodespools because updates had been published recently can be an unpleasant surprise.

[Elektronenvolt]

Some more situations where this is an issue.

• I wanted to scale down clusters before an upgrade to speed up the process
• We were running out of resources of the underlying Hyper-V cluster. I had to delete one of the small clusters to be able to continue. Scale down blocked and starting an upgrade as well (no resources). To free up resources I finally deleted a test cluster.

Set-AksHciNodePool -Name $nodepoolname -ClusterName $clu -Count $int ends up in:

I'm forced to update the cluster before I can do any scale operation