GKE Cluster immediately hits Persistent Disk SSD (GB) quota when installing Tekton files

Question

I've created a default GKE autopilot cluster and called it "tekton-cluster". After I run the following code to configure Tekton on the cluster:

# Install tekton pipelines and tasks
kubectl apply --filename https://storage.googleapis.com/tekton-releases/pipeline/latest/release.yaml

# Install tekton operators for tekton config
kubectl apply -f https://storage.googleapis.com/tekton-releases/operator/latest/release.yaml

# Install Tekton dashboard
kubectl apply --filename https://storage.googleapis.com/tekton-releases/dashboard/latest/release.yaml

# View Tekton dashboard
kubectl proxy

# Install pipelines-as-code on Kubernetes
kubectl apply -f https://raw.githubusercontent.com/openshift-pipelines/pipelines-as-code/stable/release.k8s.yaml

Nothing else is installed or running on this cluster. I can run a local Kind cluster that has all of these files and continues to work.

On GKE, however, immediately after installing all of these files, I hit the quota for Persistent Disk SSD (GB) - at 500 GB.

Also, when I run gcloud compute disks list --project=my-tekton-gcp-proj-12345, I get the following result:

Listed 0 items.

Why am I now using 500 GB of persistent disk SSD storage? This quota has been up for 2 days, despite nothing else running on the cluster right now, and no currently provisioned SSDs.

Answer 1

It might be that when the Tekton resources were installed, particularly Pipelines, it can create persistent volumes for storing build artifacts and intermediate data. And it might be that Tekton Dashboard itself might use persistent storage for caching or storing data. There could be other resources created during installation that use persistent volumes.

You can try check the Tekton resources. List all Tekton resources using kubectl get pipelines, tasks ,pipelineruns -n tekton-cluster . Identify any resources with persistent volume claims. Then review your Tekton Logs for any messages related to creating or using persistent volumes.

You can clean up unused persistent volumes. Delete unnecessary persistent claims (PVCs) or pipelines that you don't intend to use. If you need persistent storage, you can adjust Tekton resource configurations to explicitly define the desired storage size and lifecycle. Then continuously monitor your cluster's resource usage and adjust configurations to optimize storage utilization.