timeout when installing with controller-manager-0 crashloops

[karmab]

installing from openshift-v0.4.0, i get the timeout because controller-manager crashloops
a describe on the pod shows the following:

  Warning  Failed  52m (x2232 over 9h)     kubelet, r640-u10.dev1.kni.lab.eng.bos.redhat.com  Error: container create failed: container_linux.go:329: creating new parent process caused "container_linux.go:1762: running lstat on namespace path \"/proc/78318/ns/ipc\" caused \"lstat /proc/78318/ns/ipc: no such file or directory\""

furthermore, the logs of the pod indicate this error

2019/04/29 21:03:32 error listening on :8080: listen tcp :8080: bind: address already in use

indeed coredns is listening on 8080 in the corresponding node.
also i see that the node got renamed controller-manager-0, which is problematic as we need proper dns for ocp4 to be up.

[bbrowning]

What do you mean by the node got renamed? And, why would coredns listening on a node conflict with a container listening on port 8080 in a pod? 50 containers should all be able to listen on port 8080 without conflict on the same node as long as they're not trying to use a nodePort of 8080.