Feature Overview (aka. Goal Summary)  

Support upstream feature "Non-graceful node shutdown" in OCP as GA, i.e. test it and have docs for it. It is currently supported as technology preview in 4.13, see STOR-744

This feature allows Container Storage Interface (CSI) drivers to automatically detach volumes when a node goes down non-gracefully. It introduce a out-of-service node taint that an admin can set to release the volumes attachments and allows pods to be rescheduled on other nodes.

Upstream trackers:

• Enhancement issue: https://github.com/kubernetes/enhancements/issues/2268
• KEP: https://github.com/kubernetes/enhancements/pull/1116

Goals (aka. expected user outcomes)

Allow customers (admin) to release volume's attachments when a node is shutdown ungracefully.

This process is manual, there is a separate initiative that will optionally auto tag the nodes when the ungraceful shutdown nodes are detected. See ECOPROJECT-777

Requirements (aka. Acceptance Criteria):

Productise the upstream[1] feature, from tech preview to GA, test it and have docs. Optionally introduce the automation feature ECOPROJECT-777

[1]https://github.com/kubernetes/enhancements/issues/2268

Out of Scope

ECOPROJECT-777 is not considered a must have to productise the feature but would improve user experience.

Customer Considerations

Feature is currently Tech Preview, there is no change in usage. Introducing ECOPROJECT-777 would improve the user experience.

Documentation Considerations

Provide information that needs to be considered and planned so that documentation will meet customer needs.  Initial completion during Refinement status.

Interoperability Considerations

Which other projects and versions in our portfolio does this feature impact?  What interoperability test scenarios should be factored by the layered products?  Initial completion during Refinement status.