cluster-autoscaler: add ignore-instance-creation-stockout-errors

[dsafdsa1]

What type of PR is this?

/kind feature

What this PR does / why we need it:

This PR adds an option to ignore instance creation stockout errors, which would normally trigger a node group backoff and cause the scale-up to fail. Those nodes are recognized as unregistered and not unready, therefore cluster health is not affected through settings like max-total-unready-percentage and ok-total-unready-count.

Which issue(s) this PR fixes:

Fixes #

Special notes for your reviewer:

Does this PR introduce a user-facing change?

Add a `ignore-instance-creation-stockout-errors` flag.

Additional documentation e.g., KEPs (Kubernetes Enhancement Proposals), usage docs, etc.:

[k8s-ci-robot]

Hi @dsafdsa1. Thanks for your PR.

I'm waiting for a kubernetes member to verify that this patch is reasonable to test. If it is, they should reply with /ok-to-test on its own line. Until that is done, I will not automatically test new commits in this PR, but the usual testing commands by org members will still work. Regular contributors should join the org to skip this step.

Once the patch is verified, the new status will be reflected by the ok-to-test label.

I understand the commands that are listed here.

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.

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: dsafdsa1
Once this PR has been reviewed and has the lgtm label, please assign x13n for approval. For more information see the Code Review Process.

The full list of commands accepted by this bot can be found here.

Needs approval from an approver in each of these files:

• cluster-autoscaler/OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[aleksandra-malinowska]

The behaviour introduced in this PR doesn't seem related to any universal autoscaling logic, but rather "treat this VM as if it were in a different state than we (cloud provider) tell you it is because we (cloud provider) also tell you this group is special in this specific way only".

If we want to always ignore the errors from certain instances, why not modify the affected cloud provider(s) to not attach these errors to the instances from this group in the first place? When we fill instances cache, we fetch by the node group anyway, so there's easy access to this information.

I'm open to counter arguments, especially if this is a use-case that can be shared by other cloud providers, or if we use the information about stockouts in another place where we actually want to see it.

Code-wise, LGTM.