DaemonSet does not reflect current state in OpenShift Container Platform if quota exceeded

Issue

• DaemonSet does not reflect current state if quota exceeded
• We use a DaemonSet to schedule a particular pod on all compute nodes in the cluster. In case we ran into exceeded quota on the project the DaemonSet is not able to schedule pods on all desired nodes. That's reasonable and an expected behavior. But we observed that in such cases the DaemonSet does not reflect the correct state of the running pods:

$ oc get ds foo-bar
NAME         DESIRED   CURRENT   READY     UP-TO-DATE   AVAILABLE   NODE SELECTOR                          AGE
foo-bar      7         7         7         7            7           node-role.kubernetes.io/compute=true   65d

$ oc get po
NAME             READY     STATUS    RESTARTS   AGE
foo-bar-gd5tb    1/1       Running   3          48d
foo-bar-k4rhk    1/1       Running   3          48d
foo-bar-nd6jf    1/1       Running   1          30d
foo-bar-qctvf    1/1       Running   5          48d
foo-bar-vvj95    1/1       Running   1          30d
foo-bar-wvmfc    1/1       Running   1          47d