Bug 2188414

Summary:	Update Product Documentation to Reflect that all database workloads are not supported on CephFS PVs
Product:	[Red Hat Storage] Red Hat OpenShift Data Foundation	Reporter:	Craig Wayman <crwayman>
Component:	documentation	Assignee:	Kusuma <kbg>
Status:	CLOSED CURRENTRELEASE	QA Contact:	Olive Lakra <olakra>
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	unspecified	CC:	asriram, edonnell, kbg, kramdoss, odf-bz-bot, olakra
Target Milestone:	---	Keywords:	NoDocsQEReview
Target Release:	---
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2023-09-07 07:25:06 UTC	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 Craig Wayman 2023-04-20 15:59:41 UTC

Created attachment 1958626 [details]
RBD PVC Example

Describe the issue:

  We're seeing more cases in support with damaged/unrecoverable CephFS/MDS issues that usually correlate with a database application (not necessarily PostgreSQL) being hosted on a CephFS PV. Additionally, CephFS has difficulties with workloads with very high file counts (millions of small objects), even sometimes with the fsGroupChangePolicy: "OnRootMismatch" applied. 


Describe the task you were trying to accomplish:

  Would like to change the wording in the product documentation.

Suggestions for improvement:

Existing wording:

NOTE: Running PostgresSQL workload on CephFS persistent volume is not supported and it is recommended to use RADOS Block Device (RBD) volume.


Recommended Change:

NOTE:  Running any database workloads on a CephFS persistent volume is not supported (e.g. RDBMSs, NoSQL, PostgreSQL, Mongo, RocksDB, etc.) and it is recommended to use RADOS Block Device (RBD), storageclass ocs-storagecluster-ceph-rbd volumes instead (NOTE: ceph-rbd supports both block and filesystem mode). Additionally, it's strongly encouraged for applications workloads that yield high file counts (millions of small objects) to utilize ceph-rbd as well. 

Document URL:

https://access.redhat.com/documentation/en-us/red_hat_openshift_data_foundation/4.12/html-single/planning_your_deployment/index#introduction-to-openshift-data-foundation-4_rhodf

Chapter/Section Number and Title:

Chapter 1. Introduction to OpenShift Data Foundation, and scroll down to the first note.

Product Version: 
ODF v4.x

Environment Details: 
N/A

Any other versions of this document that also needs this update: 
All ODF versions.

Additional information:

Comment 3 Craig Wayman 2023-07-10 17:35:03 UTC

Good Afternoon,

  Second thought, I think it would be a good idea if we simply have this KCS[1] linked in the section of the documentation I created this doc BZ for[2].

1. https://access.redhat.com/solutions/7003415

2.  https://access.redhat.com/documentation/en-us/red_hat_openshift_data_foundation/4.12/html-single/planning_your_deployment/index#introduction-to-openshift-data-foundation-4_rhodf



Regards,


Craig Wayman
TSE Red Hat OpenShift Data Foundations (ODF) 
Customer Experience and Engagement, NA