RESOURCE_EXHAUSTED: Maximum number of keys on account reached

[Ark-kun]

deploy-ml-pipeline:	Creating a new kubernetes namespace ...
deploy-ml-pipeline:	namespace/kubeflow created
deploy-ml-pipeline:	level=info msg="Writing component at '\''/ml-pipeline-app/components/ml-pipeline.jsonnet'\''"
deploy-ml-pipeline:	Operation "operations/acf.d70b9fe1-be12-4ce0-94e8-375da4220c6d" finished successfully.
deploy-ml-pipeline:	ERROR: (gcloud.iam.service-accounts.keys.create) RESOURCE_EXHAUSTED: Maximum number of keys on account reached.
deploy-ml-pipeline:	- '\''@type'\'': type.googleapis.com/google.rpc.RetryInfo
deploy-ml-pipeline:	  retryDelay: 86401s
deploy-ml-pipeline:	secret/user-gcp-sa created

[vicaire]

Resolving as this has been inactive for several months. Please reopen if this is still an issue.

[DurivetMatthias]

I'm also getting this, this is what my bash script to deploy looks like:

CLUSTERNAME=$1
ZONE=europe-west1-b

gcloud config set compute/zone $ZONE
gcloud beta container clusters create $CLUSTERNAME \
  --enable-autoupgrade \
  --zone $ZONE \
  --scopes cloud-platform \
  --enable-cloud-logging \
  --enable-cloud-monitoring \
  --machine-type n1-standard-1 \
  --num-nodes 4

echo "cluster created"

kubectl create clusterrolebinding ml-pipeline-admin-binding --clusterrole=cluster-admin --user=$(gcloud config get-value account)

echo "clusterrole created"

# kubectl create namespace kubeflow

# kubectl create secret generic user-gcp-sa --from-file=../kubeflowScripts/user-gcp-sa.json -n kubeflow

PIPELINE_VERSION=0.1.8
kubectl create -f https://storage.googleapis.com/ml-pipeline/release/$PIPELINE_VERSION/bootstrapper.yaml

echo "pipeline jobs created"

after that the deploy-ml-pipeline job fails.
How did the previous issue get fixed?