Bug 1593418

Summary:	[Doc RFE] Tracker BZ for DFG Workgroup 3.1 documentation requirements
Product:	[Red Hat Storage] Red Hat Ceph Storage	Reporter:	Anjana Suparna Sriram <asriram>
Component:	Documentation	Assignee:	Anjana Suparna Sriram <asriram>
Status:	CLOSED CURRENTRELEASE	QA Contact:	ceph-qe-bugs <ceph-qe-bugs>
Severity:	high	Docs Contact:
Priority:	high
Version:	3.1	CC:	jharriga, jquinn, kdreyer, khartsoe, mbenjamin, mhackett, vumrao
Target Milestone:	rc
Target Release:	3.1
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:	2019-02-26 07:26:40 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:
Bug Depends On:	1593868, 1602913, 1602919, 1602921, 1602925, 1602926, 1604031
Bug Blocks:	1523244, 1581350

Description Anjana Suparna Sriram 2018-06-20 18:50:58 UTC

User Story: As a storage admin responsible for object storage, I need useful guidelines (scale limits, workload configurations, etc.) to help me design and configure an environment that is not subject to failure or performance degradation.

Master Documents: Object Gateway Guide for RHEL / Object Gateway Guide for Ubuntu

Comment 3 John Harrigan 2018-07-18 17:29:26 UTC

Updates required in "Ceph Object Gateway for Production" guide

1) TOPIC: Determining and applying expected_num_objects value
   EXPLANATION: In order to help users avoid Filestore splitting operations
                which can dramatically slow client I/O performance. While this
                behaviour can affect all Ceph users, it is especially likely to
                impact RGW customers since they likely have pools with many
                objects. Guide users through the procedure for determining the
                correct value for expected_num_objects and illustrate with
                several customer use cases.

2) TOPIC: Deploying with osd_scenario=lvm for optimal hardware usage on OSD nodes
   EXPLANATION: The installer needs to support placement of performance 
                sensitive Ceph OSD elements such as Filestore journals and RGW
                pools (i.e. bucketIndex) on NVMe devices along with OSD data
                on HDDs. Document the rationale for deploying Ceph in this
                way and the actual installation procedure for several hardware
                configurations.  

3) TOPIC: Recommended settings and procedure for applying
   EXPLANATION: There needs to a well documented procedure which details for 
                users how to apply Ceph settings. Currently users can use
                ceph-ansible, 'ceph tell/inject', directly edit ceph.conf'.
                Determine best practices, verify procedure and document.
                Provide examples for several scenarios.
 
4) TOPIC: Monitoring and controlling GC rate
   EXPLANATION: RGW garbage collection activity can adversely impact client I/O
                performance. RGW garbage collection statistics are not directly
                available to users. An explaination of GC should be provided
                along with procedures which guide users how to monitor and 
                tune RGW garbage collection.

5) TOPIC: Procedure for creating well-configured RGW pools
   EXPLANATION: RGW client I/O performance is dependent on pool settings.
                Explain the relevant parameters and document ideal pool
                creation parameters for several RGW customer use cases.

I will be opening BZs for each of these.

- John