I have an issue where instance appear in the Compute Engine → VM Instances list that when I select from the console says that it’s ‘Unable to find the resource that you requested’.
They look to be old GKE instance clusters given their names, but all the clusters in GKE are now down and the API is disabled.
Any suggestions on how I can clean these up, or even how one gets into this state?
Hi, Chris,
Can you delete the instance on your VM instances screen by checking the checkbox and selecting ‘Delete?
On your VM instances screen, does the column “In use by” appear?
It could be related to the fact that the GKE API is not enabled on the project, but that is a far-fetched theory. However, we can try to enable the API and perform the following steps:
To stop this, we can delete this instance.
Go to the Compute Engine → VM instance.
You will see the instance there, kindly click the box and select delete on the top.
Also, go to the Kubernetes Engine → Clusters screen, and if there is any cluster or pool, delete it as well.
You can disable the GKE API now.
Please let me know if this was helpful. I will be waiting for your response.
Thanks for the reply! So:
- Can you delete the instance on your VM instances screen by checking the checkbox and selecting ‘Delete?
Nope, when I do that I get the error I attached which says the instance doesn’t exist.
So while the API was still enabled I tried to remove the original cluster they were attached which errored that it was unable to remove those machines. It said they cluster would expire in a few hours so I left it to do that.
Once the cluster expired (and was no longer visible in the console) I attempted to remove the instances from the console but got the error I attached.
Finally out of desperation I disabled the GKE API but this hasn’t seemed to clean them up either.
It’s all a bit weird.
Hi, Chris,
Your project would need to be inspected by support.
This can be done over a support case.
If not, I suggest you create a Public Issue Tracker.
