Bug 1968314

We have reviewed the Operation guide sections and contents. Please find the comments below


Section 1 --> this section should come after Orchestrator and service deployments. Needs to be in sequence (mostly it should come when client is required ex: MDS, RGW, RBD) 
Upstream Doc - https://docs.ceph.com/en/latest/cephadm/client-setup/#basic-ceph-client-setup

1.2 Prerequisites - Do we need pre-requisites again despite already mentioned 1.1. 

 - "Hosts added to the cluster." - it does not related to client setup.

1.2 Procedure - 
			1. "cd /etc/ceph" -
				a. In case "/etc/ceph" directory did not exist, it should be created first on client node(node).

			2. Generate the configuration file -
				a. It should be very clear about, where to execute this command to avoid confusion between 1.2.1 and 1.2.2 .
				b. After getting the config, where it should be stored?

				Upstream Doc - https://docs.ceph.com/en/latest/cephadm/client-setup/#config-file-setup

1.3. SETTING-UP KEYRING ON CLIENT MACHINES - Do we need pre-requisites again
			1. "cd /etc/ceph" -
				a. In case "/etc/ceph" directory did not exist, it should be created first on client node(node).

Section 2. Generate the keyring for the client:
				a. It should be very clear about, where to execute this command to avoid confusion.
				b. After getting the Keyring, where it should be stored and what should be the file names?

			https://docs.ceph.com/en/latest/cephadm/client-setup/#keyring-setup

2.2. OVERVIEW OF RED HAT CEPH STORAGE ORCHESTRATORS

		Rook Orchestrator - QE did not qualify this, not sure this should be part of doc.


2.3. MANAGEMENT OF SERVICES USING THE CEPH ORCHESTRATOR

		
        1.Example provided for Using specification yaml file is incorrect. 
		2. It would be to use same MON example mentioned in placement specification through CLI.

	2.3.2, 2.3.3, 2.3.4, 2.3.5, 2.3.6 explains same but does not seems to be in order which sould be understood easily.

2.4 - Monitoring stack services - Should be in order after host add add, mon, mgr, OSD deployments


2.5(Host management) - should come first in order before 2.4(Monitoring stack services).

2.6 - Management of monitor daemons using the Ceph Orchestrator ( Need to use service instead of daemons in the headline - Check with Dev on this)

2.7 - Management of manager daemons using the Ceph Orchestrator - ( Need to use service instead of daemons in the headline - Check with Dev on this)

2.8 - Management of OSD daemons using the Ceph Orchestrator - ( Need to use service instead of daemons in the headline - Check with Dev on this)

2.11 -NFS ganesha section for RGW and CEPHFS is missing.  We should have steps for NFS over RGW and NFS over CephFS



Section 3: Ceph OSDs --Needs to be revisited - and removed from the contents whichever is not applicable as i see creation/removal of OSDs using ansible etc..ceph-volume is not supported in 5.0 - Content needs to be revisited

Section 4 - Replace OSDs -- Needs to be  w.r.t 5.0 commands using orch - Content needs to be verified again and retain and add whatever is applicable for 5.0


Section 5 and 6  - Handlink node failure and data center failure - Using ansible steps are provided - Needs to be revisited for both the section - Add/retain whatever is applicable for 5.0 as we see Add/remove using ansible which is not supported for 5.0