Description of problem:
-----------------------
Update from 4.9.29 to 4.10.13 stuck and cluster's api unresponsive:

[kni@provisionhost-0-0 ~]$ oc get mcp
The connection to the server api.edge-0.qe.lab.redhat.com:6443 was refused - did you specify the right host or port?

[kni@provisionhost-0-0 ~]$ oc get nodes
The connection to the server api.edge-0.qe.lab.redhat.com:6443 was refused - did you specify the right host or port?

[kni@provisionhost-0-0 ~]$ oc get co
The connection to the server api.edge-0.qe.lab.redhat.com:6443 was refused - did you specify the right host or port?

After checking nodes via console found that IP addresses on 2 out of 3 nodes were on `bond0` interface as well as on `br-ex`:
===========================================
[root@master-0-0 ~]# ip a s bond0
11: bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 1500 qdisc noqueue master ovs-system state UP group default qlen 1000
    link/ether 52:54:00:5e:3a:cc brd ff:ff:ff:ff:ff:ff
[root@master-0-0 ~]# ip a s br-ex
14: br-ex: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default qlen 1000
    link/ether 52:54:00:5e:3a:cc brd ff:ff:ff:ff:ff:ff
    inet 192.168.123.123/24 brd 192.168.123.255 scope global dynamic noprefixroute br-ex
       valid_lft 2933sec preferred_lft 2933sec
    inet6 fe80::5054:ff:fe5e:3acc/64 scope link noprefixroute
       valid_lft forever preferred_lft forever
[root@master-0-1 ~]# ip a s bond0
11: bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether 52:54:00:7a:77:cc brd ff:ff:ff:ff:ff:ff
    inet 192.168.123.64/24 brd 192.168.123.255 scope global dynamic noprefixroute bond0
       valid_lft 549sec preferred_lft 549sec
    inet 192.168.123.5/32 scope global bond0
       valid_lft forever preferred_lft forever
    inet 192.168.123.10/32 scope global bond0
       valid_lft forever preferred_lft forever
[root@master-0-1 ~]# ip a s br-ex
15: br-ex: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default qlen 1000
    link/ether 52:54:00:a6:95:c6 brd ff:ff:ff:ff:ff:ff
    inet 192.168.123.92/24 brd 192.168.123.255 scope global dynamic noprefixroute br-ex
       valid_lft 3273sec preferred_lft 3273sec
    inet6 fe80::5054:ff:fea6:95c6/64 scope link noprefixroute
       valid_lft forever preferred_lft forever
[root@master-0-2 ~]# ip a s bond0
11: bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether 52:54:00:da:64:62 brd ff:ff:ff:ff:ff:ff
    inet 192.168.123.79/24 brd 192.168.123.255 scope global dynamic noprefixroute bond0
       valid_lft 2636sec preferred_lft 2636sec
    inet 192.168.123.5/32 scope global bond0
       valid_lft forever preferred_lft forever
    inet 192.168.123.10/32 scope global bond0
       valid_lft forever preferred_lft forever
[root@master-0-2 ~]# ip a s br-ex
14: br-ex: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default qlen 1000
    link/ether 52:54:00:52:7e:73 brd ff:ff:ff:ff:ff:ff
    inet 192.168.123.106/24 brd 192.168.123.255 scope global dynamic noprefixroute br-ex
       valid_lft 2418sec preferred_lft 2418sec
    inet6 fe80::5054:ff:fe52:7e73/64 scope link noprefixroute
       valid_lft forever preferred_lft forever


Version-Release number of selected component (if applicable):
=============================================================
4.10.13


How reproducible:
=================
from time to time

Steps to Reproduce:
===================
1. Deploy baremetal IPI cluster with bonds configured on 'baremetal' network
2. Start OCP update


Actual results:
===============
Update stuck; wrong interfaces got IP addresses assigned


Expected results:
=================
Update succeeds; expected interfaces got IP addresses assigned

Additional info:
================
Virtual setup: 3 masters + 4 worker nodes