Bug 2129928 - [MetroDR] Deploy 2 admin nodes (one in each zone) instead of one in a stretched ceph cluster
Summary: [MetroDR] Deploy 2 admin nodes (one in each zone) instead of one in a stretch...
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: None
Product: Red Hat OpenShift Data Foundation
Classification: Red Hat Storage
Component: documentation
Version: 4.11
Hardware: Unspecified
OS: Unspecified
unspecified
medium
Target Milestone: ---
: ---
Assignee: Olive Lakra
QA Contact: akarsha
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2022-09-26 16:39 UTC by Martin Bukatovic
Modified: 2023-08-09 16:43 UTC (History)
5 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2023-03-09 12:47:11 UTC
Embargoed:


Attachments (Terms of Use)

Description Martin Bukatovic 2022-09-26 16:39:41 UTC
Describe the issue
==================

Section "Deploying Red Hat Ceph Storage" in Metro DR part of the ODF DR guide
instructs admin to deploy *only one* admin node in the cluster. This means that
in case of a zone failure which includes this admin node, there is no other
admin node readily available to quickly check status of ceph cluster.

Describe the task you were trying to accomplish
===============================================

Deploy and configure stretched ceph cluster for Metro DR use case, so that
after losing one zone/datacenter, admin can quicly check ceph status on
an admin node of operational zone.

Suggestions for improvement
===========================

Instead of a single admin node, select 2 such nodes, one in each
zone/datacenter.

This change will involve:

- the ansible inventory (the admin group there will need to include 2 nodes,
  one from each zone)
- example of `ceph orch host ls` command will need to include 2 nodes with
  _admin label

Document URL
============

https://access.redhat.com/documentation/en-us/red_hat_openshift_data_foundation/4.11/html/configuring_openshift_data_foundation_disaster_recovery_for_openshift_workloads/metro-dr-solution

Chapter/Section Number and Title
================================

Product Version
===============

ODF 4.11

Environment Details
===================

Metro DR with a stretched ceph cluster

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

I don't think so.

Additional information
======================

Based on experience with testing negative scenarios and suggestion
from #rh-ceph channel.

Comment 1 Martin Bukatovic 2022-12-16 17:17:25 UTC
I'm checking whether the following changes will address the request proposed in this bug:

- add ceph4 node into "admin" group in ansible inventory
- add "_admin" label to entry of ceph4 node in cluster-spec.yaml file

Comment 8 Harish NV Rao 2023-02-17 09:17:49 UTC
Targeting this to 4.12.1. QE is already testing with this config


Note You need to log in before you can comment on or make changes to this bug.