Instructions and Helm Charts for containers in use are confusing and provide different behaviours

[saintdle]

/kind bug

What happened?

When following the linked instructions for snapshot feature with the EBS CSI driver, creating a snapshot just hangs completely. Unable to read any logs from the snapshot controller as it's exited and no previous logs are available.

If I roll the deployment back, and deploy the CSI Driver using Helm with the feature gates. And then annotate my controller pods with ARN, and install the CRDs from here ,as there are none deployed by Helm. Then things work as expected.

What you expected to happen?

When following the standard install it should work with the external pre-reqs links provided.

When using helm, the snapshot related CRDs should be installed

How to reproduce it (as minimally and precisely as possible)?

Using the out of the box instructions - deploy driver with alpha features

kubectl apply -k "github.com/kubernetes-sigs/aws-ebs-csi-driver/deploy/kubernetes/overlays/alpha/?ref=master"

Then follow this provided snapshot instructions link. When you get to Step 3 of the pre-reqs in this page, you are sent to this document. Once followed those steps to create CRD, and snapshot controller. And then create the VolumeSnapshotClass from the provided example back on the AWS EBS CSI snapshot page.

When running the example pv and snapshot creation. The PV creates ok, but the snapshot just says;

Events:
Type Reason Age From Message

---

Normal CreatingSnapshot 5m36s snapshot-controller Waiting for a snapshot default/new-snapshot-demo to be created by the CSI driver.

Anything else we need to know?:

Environment

• Kubernetes version (use kubectl version): v1.18.9-eks-d1db3c
• Driver version: latest

[wongma7]

Sorry about this, yes we're aware of this issue...by default, the helm chart doesnt install necessary snapshot, including CRD, and it's not documented that you need to do it yourself. I think this is a consequence of the chart not being updated for a while (since snapshot feature was alpha but has since graduated to ga... not making excuse, just explaining.)

related:
#797

[saintdle]

Sorry about this, yes we're aware of this issue...by default, the helm chart doesnt install necessary snapshot, including CRD, and it's not documented that you need to do it yourself. I think this is a consequence of the chart not being updated for a while (since snapshot feature was alpha but has since graduated to ga... not making excuse, just explaining.)

related:
#797

No worries, was just a good few hoops to jump through, and we kept hitting different outcomes for what we needed at the time.

[krmichel]

This issue is related to issue #635

[krmichel]

Version 2.0.0 of the chart no longer installs the snapshot controller or its CRDs. The documentation linked here should be updated to reflect that.

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue or PR as fresh with /remove-lifecycle stale
• Mark this issue or PR as rotten with /lifecycle rotten
• Close this issue or PR with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale