Recently applied this fix for controllers due to the systemd-journald restart issue, it sounds as is this one is also sometimes affected. Containerized already uses this, so we will apply the same restart strategy here.
@@ -15,6 +15,7 @@ LimitNOFILE=131072
LimitCORE=infinity
WorkingDirectory={{ openshift.common.data_dir }}
SyslogIdentifier=atomic-openshift-master-api
+Restart=always
RestartSec=5s
[Install]