Numaresources Operator version 4.18 is incorrectly installed on OpenShift 4.17 clusters

Solution Verified - Updated 19 Feb 2026

Environment

Red Hat OpenShift Container Platform (RHOCP)
- 4.17
Numaresources Operator 4.18

Issue

After a catalog update, the Numaresources Operator may have been upgraded to version 4.18, which is not supported on OpenShift Container Platform (OCP) versions 4.17.

This article covers the specific remediation for the Numaresources Operator. For the general issue affecting multiple operators, refer to Red Hat Operator has version higher than the cluster version.

Symptoms

Administrators can see that the 4.18 Numaresources operator is installed:

$ oc get sub,csv -n openshift-numaresources

NAME                                                                              PACKAGE                  SOURCE             CHANNEL
subscription.operators.coreos.com/openshift-numaresources-operator-subscription   numaresources-operator   redhat-operators   4.18

NAME                                                                        DISPLAY                  VERSION   REPLACES                         PHASE
clusterserviceversion.operators.coreos.com/numaresources-operator.v4.18.2   numaresources-operator   4.18.2    numaresources-operator.v4.17.5   Succeeded

Resolution

1. Preconditions & Safety Checks

Before initiating the removal or installation process, the following conditions must be met:

Operator version mismatches the cluster’s: numaresources operator is found in a mismatching installed version to the OCP cluster, for example: a 4.17 OCP cluster has 4.18 operator.
No Active Operations: Verify that no CGU (Cluster Group Upgrades) or IBU (Image Based Upgrades) are currently in progress.
Backup Policy: The NROP operator controller manager alone isn’t sufficient enough for the recovery to work properly after the fix, the user must backup the numaresourcesoperator and numaresourcescheduler instances (if existed) in order to restore the old configuration, or once the controller has been reconfigured, re-apply the instances.

2. Phase 1: Clean Uninstallation (CLI Commands)

To completely remove the operator and its metadata, execute the following commands in sequence:

Check the current version of the Subscription and CSV:

oc get subscription -n openshift-numaresources
oc get csv -n openshift-numaresources

Delete the subscription

oc delete subscription numaresources-operator -n openshift-numaresources

# Expected output: subscription.operators.coreos.com "numaresources-operator" deleted

Delete the Cluster Service Version (CSV):
Replace 4.18.0 with the specific version string found in step 1.
```
oc delete csv numaresources-operator.v4.18.0 -n openshift-numaresources
```

Remove Custom Resource Definitions related to the NROP Operator
Note: Only remove CRDs if a fresh state is required for the new installation.
By removing the CRD’s we are removing the NROP instance which will cause a reboot to all of the corresponding nodes that the CR was pointing to.

oc get crd | grep -i numaresources
# Expected output:
# numaresourcesoperators.nodetopology.openshift.io                  
# numaresourcesschedulers.nodetopology.openshift.io

oc get crd | grep -i noderesourcetopologies
# Expected output:
# noderesourcetopologies.topology.node.k8s.io

oc delete crd numaresourcesoperators.nodetopology.openshift.io numaresourcesschedulers.nodetopology.openshift.io noderesourcetopologies.topology.node.k8s.io

Here after deleting the CRD’s we need to wait for the corresponding Nodes and MCP to return back to a ready state before starting with phase 2.

3. Phase 2: Targeted Reinstallation

To reinstall and pin the operator to the desired version, create a Subscription manifest pointing to the version required that aligns with your OCP cluster’s version, and with a manual installPlan approval.

Subscription Manifest (`nrop-subscription.yaml`)

apiVersion: [operators.coreos.com/v1alpha1](https://operators.coreos.com/v1alpha1)
kind: Subscription
metadata:
  name: numaresources-operator
  namespace: openshift-numaresources
spec:
  channel: <version> # "4.17" for example
  name: numaresources-operator
  source: redhat-operators
  sourceNamespace: openshift-marketplace
  installPlanApproval: Manual

Apply the Installation:

Create the subscription:
```
oc apply -f nrop-subscription.yaml
```

Approve the InstallPlan:
Since installPlanApproval is set to Manual, you must retrieve and approve the plan:

oc get installplan -n openshift-numaresources
oc patch installplan <installplan_name> -n openshift-numaresources --type merge --patch '{"spec":{"approved":true}}'

4. Post-Installation Validation

Important Note: If the CRDs were removed in Phase 1.4, we need to reapply the backed-up YAML files here, which will trigger a MCP update, and the affected nodes will reboot.

Verify Operator Status:
```
oc get csv -n openshift-numaresources
```
Verify Pods:
```
oc get pods -n openshift-numaresources
```

Root Cause

For several hours on February 3, 2026, Red Hat released 4.18 Red Hat Operators catalog content into 4.12-4.17 clusters.

Diagnostic Steps

For more information, including Diagnostic Steps, please refer to Red Hat Operator has version higher than the cluster version.

This solution is part of Red Hat’s fast-track publication program, providing a huge library of solutions that Red Hat engineers have created while supporting our customers. To give you the knowledge you need the instant it becomes available, these articles may be presented in a raw and unedited form.