GCP Compute Engine - Persistent Disk SSD Quota Reached Issue

Question

After trying to create a new Kubernetes cluster which was configured to require 300GB of persistent disk space (1 zone * 3 nodes * 100GB boot disk size), but ultimately failed to build due to the cluster deploy target zone not having enough resources, I am now trying deploy a new cluster with the same settings in a different zone after deleting the erring cluster from my previous attempt.

However, my Persistent Disk SSD quota for Compute Engine now says I have reached the limit I currently have set (500GB) when, in fact, I currently have no Compute Engine instances listed in my console and no disks to show after running $ gcloud compute disks list in my Cloud Shell Editor.

What am I missing here?

A few hypothesis I have:

1. Is it possible that there was some sort of hang-up issue caused by my first attempt and now I am using disk space on some nodes that I no longer have a means to access?
2. Is this just the result of some sort of server-side caching issue with quotas? (though I should mention it has been several hours since this quota limit was reached and it still shows it is at the limit)
3. Is the Persistent Disk SSD quota a temporal quota (e.g. 500GB allocated per month) as opposed to the way it reads ~ "500GB is the total disk space you can occupy across nodes at any given moment"?

Any input as to the root of this problem would be appreciated.

Answer 1

It turns out that either my second hypothesis was correct, and there was some sort of caching related delay, or whatever task(s) that are responsible for deallocating clusters and disk space take quite some time (hours) to complete and update the quota usages. I suspect the latter is true. I'll leave this question/answer up in case anyone else may find it useful.