My customer is reporting the following issue:

We have made a pipeline to deploy an OpenShift Cluster which was working fine until a fews weeks ago. 

When we deploy a new OpenShift cluster we see that 30% of the nodes (worker, storage and infra) are in a endless loop: "Get error: Get 'https://api-int.xxx.x.xxx.xx:22623/config/worker' : dial tcp: lookup api-int.xxx.xxxx.xxxx.xx on [::1]:53: read udp [::1]:35946->[::1]:53: read: connection refused"
This message will scroll endlessly over the console. See attachment of console screenshot.

When we give a "ctrl-alt-del" in the console of the hanging node, the node will boot and continue successfully. The installation will complete okay for that node.

4.11.26

Difficult and random

1.
2.
3.

Some nodes get stuck at first boot

All nodes to be deployed without issue