Assigned
Status Update
No update yet.
Comments
jf...@streamdata.io
Hello,
Thanks for reaching out to us!
The Product Engineering Team has been made aware of your feature request, and will address it in due course. Though we can't provide an ETA on feature requests nor guarantee their implementation, rest assured that your feedback is always taken very seriously, as it allows us to improve our products. Thank you for your trust and continued support to improve Google Cloud Platform products.
In case you want to report a new issue, please do not hesitate to create a new [Issue Tracker]
Thanks and Regards,
Onkar Mhetre
Google Cloud Support
ju...@google.com
ju...@google.com
Specifically, this happened to us. We have a GKE cluster with two similarly-sized node pools. One has E2 nodes and the other has C2D nodes. We wanted to temporarily double the size of our E2 nodepool and checked the quotas page first. The only quota in the entire GCP project that was above 20% was "CPUs" at 55%. We reasoned that since only half of our machines were in the nodepool we wanted to double, this would only raise us to around 80-85% of quota, which seemed safe for a temporary change. We were surprised to see our scale-up fail due to quota. (And then we clumsily tried to quickly scale down to get below quota using direct GCE ASG scaling rather than GKE autoscaler, which ended up deleting a bunch of our more active machines instead of the empty ones — perhaps user error, but a bigger impact than we had expected for what we thought was going to be a simple "scale up fast, let it scale down slowly once our temporary need was over".)
Had the human-readable label in the UI said "E2 and N1 CPUs" instead of "CPUs" we would not have made that mistake: we would have tried for a smaller scale-up or requested more quota first.
I assume that the "internal"/computer-readable name `CPUS` is unlikely to change for compatibility reasons, but hopefully the UI display name is less hardcoded?
Had the human-readable label in the UI said "E2 and N1 CPUs" instead of "CPUs" we would not have made that mistake: we would have tried for a smaller scale-up or requested more quota first.
I assume that the "internal"/computer-readable name `CPUS` is unlikely to change for compatibility reasons, but hopefully the UI display name is less hardcoded?
pd...@gmail.com
FWIW, this issue is still current and often bites when managing service accounts through the API, for example using terraform.
[Deleted User] <[Deleted User]>
We just spent 2 days chasing a ghost where the root cause was not idempotent service account id. It would be extremely helpful to fix this bug.
Description
When this service account is deleted, the IAM role is not automatically deleted.
The feature request is to automatically delete the IAM role that was automatically created.
The reason for this is that recreating a service account with the same name after it has been deleted will merge permissions for old and new one in IAM role without necessarily being aware of it.
Thanks