-
Feature Request
-
Resolution: Unresolved
-
Undefined
-
None
-
None
-
None
-
False
-
None
-
False
-
Not Selected
-
-
-
-
1. Proposed title of this feature request
Add a way to update (automatic or manual) infrastructureTopology cluster parameter
2. What is the nature and description of the request?
Currently, the parameter infrastructureTopology in the
infrastructure.config.openshift.io
is set at installation time [1]:
- no workers -> infrastructureTopology = controlPlaneTopology
- 1 worker -> infrastructureTopology = SingleReplica
- 2 or more workers -> infrastructureTopology = HighlyAvailable
Many of the clusters are installed with 3 masters and 1 worker (as it's enough to install a cluster), and then workers are added by scaling up the worker machineset as a day-2 operation based on our needs.
The problem is that infrastructureTopology is used by some cluster operators, like monitoring (CMO), to determine the number of replicas or the scrape_interval [2]. But in Openshift clusters with a dynamic number of nodes, it makes no sense at all to have a static infrastructureTopology based on the installation, when the infrastructureTopology can change in day-2.
[2] https://github.com/openshift/cluster-monitoring-operator/pull/1652
Customer would like a way to change this parameter (can be done in a cluster operator, or as a parameter that can be edited).
3. Why does the customer need this? (List the business requirements here)
They have many clusters affected by this, and currently is not supported to change the infrastructureTopology.
4. List any affected packages or components.
installer
- is related to
-
OCPBUGS-25868 The infrastructureTopology should be updated to `HighlyAvailable` after the worker nodes scale up to 2
- New
- relates to
-
RFE-5489 Customize cluster infrastructureTopology after cluster installation
- Backlog