corosync service fails to start during node boot in a cloud environment with RRP configured

Issue

• Cluster nodes during boot can't start corosync.service with Redundant Ring Protocol(RRP) and dhcp set

• The corosync.service unit fails, and the following error messages are observed in the logs: 

  May 11 08:00:01 node1 corosync[1158]:  [MAIN  ] Corosync Cluster Engine ('2.4.5'): started and ready to provide service.
  ...
  May 11 08:00:01 node1 corosync[1340]:  [TOTEM ] adding new UDPU member {1.x.x.33}
  May 11 08:00:01 node1 corosync[1340]:  [TOTEM ] adding new UDPU member {1.x.x.34}
  May 11 08:00:01 node1 corosync[1340]:  [TOTEM ] One of your ip addresses are now bound to localhost. 
  Corosync would not work correctly.