When deploying a whereabouts-IPAM-based additional network through the cluster-network-operator, the whereabouts-reconciler daemonset is not deployed on non-amd64 clusters due to an hard-coded nodeSelector introduced by https://github.com/openshift/cluster-network-operator/commit/be095d8c378e177d625a92aeca4e919ed0b5a14f

4.13+

Always. Tested on a connected arm64 AWS cluster using the openshift-sdn network

1. oc new-project test1
2. oc patch networks.operator.openshift.io/cluster -p '{"spec":{"additionalNetworks":[{"name":"tertiary-net2","namespace":"test1","rawCNIConfig":"{\n  \"cniVersion\": \"0.3.1\",\n  \"name\": \"test\",\n  \"type\": \"macvlan\",\n  \"master\": \"bond0.100\",\n  \"ipam\": {\n    \"type\": \"whereabouts\",\n    \"range\": \"10.10.10.0/24\"\n  }\n}","type":"Raw"}],"useMultiNetworkPolicy":true}}' --type=merge
3. oc get daemonsets -n openshift-multus 

NAME                            DESIRED   CURRENT   READY   UP-TO-DATE   AVAILABLE   NODE SELECTOR              AGE
whereabouts-reconciler          0         0         0       0            0           kubernetes.io/arch=amd64   7m27s

No kubernetes.io/arch=amd64 set, so that non-amd64 and multi-arch compute clusters can schedule the daemonset on each node, regardless of the architecture.

Same problem on s390x