OpenShift Container Platform upgrade is missing serial restart for Control-Plane during API aggregation configuration

Issue

• We applied the Security patch for OpenShift (Update from 3.7.64 to 3.7.72). For this upgrade we used the recommended way with the provided ansible playbooks (playbooks/byo/openshift-cluster/upgrades/v3_7/upgrade_control_plane.yml). Now the problem is, that at some point, this playbook restarts all master api at the same point in time which is not ok for production workloads.

2019-02-01 06:31:56,541 p=94084 u=foo |  TASK [restart master api] 
2019-02-01 06:32:10,904 p=94084 u=foo |  changed: [10.10.10.1]
2019-02-01 06:32:11,688 p=94084 u=foo |  changed: [10.10.10.2]
2019-02-01 06:32:16,878 p=94084 u=foo |  changed: [10.10.10.3]

2019-02-01 06:32:16,944 p=94084 u=foo |  TASK [restart master controllers] 
2019-02-01 06:32:19,285 p=94084 u=foo |  changed: [10.10.10.1]
2019-02-01 06:32:19,341 p=94084 u=foo |  changed: [10.10.10.2]
2019-02-01 06:32:21,150 p=94084 u=foo |  changed: [10.10.10.3]