Description of problem:

https://github.com/redhat-developer/rhdh-operator/pull/1807/files modified the RHDH CSV by adding this label:

operatorframework.io/arch.arm64: supported

And it also removed the amd64 matchExpression from the node affinity.

See:
https://github.com/redhat-developer/rhdh-operator/blob/main/bundle/rhdh/manifests/backstage-operator.clusterserviceversion.yaml#L66

The problem is that the label is used in the OCP console to filter out which installable operators to list: https://github.com/openshift/console/blob/main/frontend/packages/operator-lifecycle-manager/src/components/operator-hub/operator-hub-items.tsx#L102-L139

Adding arm64 will make the RHDH Operator show up on ARM64 clusters, which would falsely assume support for such architectures, which is not (yet?) the case, at least for the downstream operator image built via Konflux.
Right now, the RHDH Operator (1.8 GA) doesn't show up at all on ARM64 clusters, which is expected:

And the node affinity prevents K8s from scheduling the controller pod on a non-compatible Node. So removing it makes it potentially scheduled on an arm64 node.

Prerequisites (if any, like setup, operators/versions):

ARM64 OCP cluster

Steps to Reproduce

 # <steps>

Actual results:

Expected results:

While we can keep the updates from the PR above in the upstream CSV, we should remove it in the downstream CSV. Otherwise, users might be able to install the operator, but the controller container will end up crashing due to wrong node architecture.

Reproducibility (Always/Intermittent/Only Once):

Always

Build Details:

OCP 4.19