Issue

• I changed node configuration as described in documentation, added to the kubelet arguments, and config file is found and read

kubeletArguments:
  cloud-provider:
    - "openstack"
  cloud-config:
    - "/etc/origin/node/cloud.conf"

• Change node config, restart node.

• Logs go into a loop:

Jul 12 20:08:00 node1.example.com atomic-openshift-node[98369]: I0712 20:08:00.243409   98435 kubelet.go:2770] Recording NodeHasSufficientDisk event message for node node1.example.com
Jul 12 20:08:00 node1.example.com atomic-openshift-node[98369]: I0712 20:08:00.243459   98435 kubelet.go:1134] Attempting to register node node1.example.com
Jul 12 20:08:00 node1.example.com atomic-openshift-node[98369]: E0712 20:08:00.822027   98435 kubelet.go:1173] Previously "node1.example.com" had externalID "[id]"; now it is "node1.example.com"; will delete and recreate.
Jul 12 20:08:00 node1.example.com atomic-openshift-node[98369]: E0712 20:08:00.823340   98435 kubelet.go:1175] Unable to delete old node: User "system:node:node1.example.com" cannot delete nodes at the cluster scope

• Updated as instructed via docs

Status of nodes prior to change:

NAME                      STATUS                     AGE
node1.example.com   Ready,SchedulingDisabled   23h
node2.example.com   Ready                      23h
node3.example.com   Ready                      23h

Status after node restarted

AME                      STATUS                        AGE
node1.example.com   NotReady,SchedulingDisabled   1d
node2.example.com   Ready                         1d
node3.example.com   Ready                         1d

Removed changes to node configuration, restarted both master and node and status back to Ready.