We have our own Registry (like dockerhub) from where we can pull images. (the registry is in the same private network 192.168.25.x).
Now we're trying to install OpenShift (very basic: 1 master + 1 node) on 192.168.25.1 and 192.168.25.2.
We have experience with those installs but than we used public ip's.
We have SSH acces from our master to our node.
But: prereqs: you need ansible on the master, git, docker on master and node, ...
- Do we need a public IP on every instance when we want to run the playbook? (it failed for resolving something to check ik yum-utils were installed).
- Is this a good solution?: (public IP and private on master and node). Install prereqs and execute playbook. So we have a cluster. After that deleting the public network and reexecuting the playbook with only private ip's (or only a public ip on the master). Will this work?
So as you can see we can use some input in using the setup.
If someone has experience with the setup of OpenShift where the communication over routes (through the router) happends internally (so no public wildcard). Please share some knowledge :).