Bug 2190049

Summary: [DOC] ODF documentation should clearly describe what creating a storage class with a Reclaim Policy of "Retain" implies
Product: [Red Hat Storage] Red Hat OpenShift Data Foundation Reporter: bmcmurra
Component: documentationAssignee: Anjana Suparna Sriram <asriram>
Status: NEW --- QA Contact: Neha Berry <nberry>
Severity: medium Docs Contact:
Priority: unspecified    
Version: unspecifiedCC: odf-bz-bot
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description bmcmurra 2023-04-26 23:17:06 UTC
Describe the issue:

Our documentation[1] does a poor job of explaining that PVCs provisioned under a Storage class that has a  Reclaim Policy of "Retain" requires the administrator to manually delete volumes from the Ceph's backend if they want to fully delete a PVC and its contents.

Describe the task you were trying to accomplish:

Improve our documentation to describe that "manual cleanup" of a Retained PVC requires that the administrator must delete the orphaned rbd/cepfs volumes directly from the Ceph backend.

Suggestions for improvement:

Something more along the lines of kubernetes upstream documentation on this matter[2]. Better would be using something along the lines of [2], but adding in the fact that any attempt at cleaning up PVCs that were provisioned with a StorageClass with a "Retain" Reclaim Policy will require the admin to login to the rook-ceph operator or rook-ceph tools pod, and remove the rbd/cephfs volume manually using Ceph commands that we don't document publicly.

Document URL:

https://access.redhat.com/documentation/en-us/red_hat_openshift_data_foundation/4.9/html/managing_and_allocating_storage_resources/storage-classes_rhodf
Chapter/Section Number and Title:

Product Version:

ODF 4.x

Environment Details:

ODF 4.x

Any other versions of this document that also needs this update:

Additional information:

Justification: I've seen multiple customers hold the misconception that PVCs provisioned with a StorageClass that has Reclaim Policy of "Retain" can be fully cleaned up simply by deleting the PVC (i.e. the same way one would clean up PVCs with a Reclaim Policy of "Delete".

This has lead to many instances of us (support) having to a) educate the customer on what a Retention reclaim policy implies, and b) cleaning up their orphaned rbd/cephfs volumes manually for them.

I think this misconception can be avoided simply by making our documentation clearer on the different Reclaim Policies and their implications.

[1] https://access.redhat.com/documentation/en-us/red_hat_openshift_data_foundation/4.9/html/managing_and_allocating_storage_resources/storage-classes_rhodf

[2] https://kubernetes.io/docs/concepts/storage/persistent-volumes/