Description of problem:
After a new installation of an OCP 3.4 there is a strange behaviour of the deployment pods.
Randomly the deployer pod fails to run with the following error:

oc logs busybox-1-deploy
--> Scaling busybox-1 to 1
error: couldn't scale busybox-1 to 1: Scaling the resource failed with: replicationcontrollers "busybox-1" is forbidden: not yet ready to handle request; Current resource version 1868422
[root@ocpmastprd1 ~]# oc get po -o wide
NAME               READY     STATUS    RESTARTS   AGE       IP             NODE
busybox-1-deploy   0/1       Error     0          1m        10.240.6.76    ocpnodinfprd1.spb.lan

We have found out that it is not related to the node itself as some deployments have worked for the same image and on the same node.
When the deploy pod succeeds then all the application pods are running without problems.

Maybe it is related to the ose-deployer image or to the docker configuration. I don't know how to gather more information.

I have been addressed to this kubernetes issue, but I don't know if it is related: https://github.com/kubernetes/kubernetes/issues/35068

Version-Release number of selected component (if applicable):
oc v3.4.1.7
kubernetes v1.4.0+776c994
features: Basic-Auth GSSAPI Kerberos SPNEGO

How reproducible: Unknown


Steps to Reproduce:
1. 
2.
3.

Actual results:
The deploy fails

Expected results:


Additional info: