EKS Storage Volume Support about helm-charts

[Crazybus]

I work at Nike and we are using EKS.

While we don't have any automated testing yet for EKS it has been manually confirmed to work. As long as the Kubernetes provider has some kind of persistent storage (which EKS does) it's going to work just fine.

Have you been able to run other statefulsets with the same volumeClaimTemplate configuration?

Could you give me the following to help debug:

• Your values.yaml configuration (mostly interested in the volumeClaimTemplate settings).
• Output of the following commands:

kubectl get sc
kubectl get pvc
kubectl get events
kubectl get sts elasticsearch-master -o yaml # (or whatever your deployment is called)
kubectl describe pod elasticsearch-master-0

[NEPDAVE]

output.txt

Thank you for getting back to me! It is much appreciated. We tried a workaround that fixed the original error message. Not sure if this is a "real" workaround, dumb luck, or something else. I have also attached the requested output. The output is from the current configuration which is not encountering the original error message.

Again, here is the original error message:
pod has unbound immediate PersistentVolumeClaims (repeated 3 times)

The steps taken for the "workaround" are here:

We created a new storage class on the cluster named "elasticsearch" of type "gp2".

Her is the yaml:

kind: StorageClass
apiVersion: storage.k8s.io/v1
metadata:
  name: elasticsearch
  annotations:
    storageclass.kubernetes.io/is-default-class: "true"
provisioner: kubernetes.io/aws-ebs
parameters:
  type: gp2
  fsType: ext4 

In the values.yaml file we changed the storageClassName: "standard" variable to storageClassName: "elasticsearch".

Here is the volumeClaimTemplate section taken from loading the vaules.yalm variables into the statefulset.yaml template:

volumeClaimTemplates:
  - metadata:
      name: elasticsearch-master
    spec:
      accessModes:
      - ReadWriteOnce
      resources:
        requests:
          storage: 30Gi
      storageClassName: elasticsearch

Deploying this configuration via Helm worked. We then deleted the resources from the "workaround" configuration and tried going back to our "original" configuration - IE we changed the values.yaml file storageClassName: "standard" variable to storageClassName: "gp2" . and deployed again via Helm and this also worked.

Here is the volumeClaimTemplate we currently have in the cluster:

volumeClaimTemplates:
  - metadata:
      name: elasticsearch-master
    spec:
      accessModes:
      - ReadWriteOnce
      resources:
        requests:
          storage: 30Gi
      storageClassName: gp2

We are going to continue testing this for a proof of concept. Thank you for your help. Please let me know if you have any questions or require additional information.

[NEPDAVE]

I am going to close the issue seeing how we are not encountering the original error message. Again, please let me know if you have any follow up questions or require additional information.