So Jeff has gone to a better place. My many year old ATOM that I was using as one of the nodes and bootstrap server for OpenShift 3.11. Its CPU was clocked at 80’c and then overnight it never returned.

Jeff held my inventory file (config file) for my OpenShift and with the passing off Jeff (and his disk) I have lost the file and all notes I had made.

The purpose of this series of Articles is to document my install process of OKD on Fedora 29 (Linux distro). OKD is the OpenSource version of OpenShift. I want OKD over K8s or K3s because it is the K8s platform of choice by all big businesses. However as I have stated many many times before I am not an infra guy. I am happier at the application layer and above.

I will document hopefully using ASCIINEMA everything I do in the hope that it will help someone else hitting these same issues.

My environment will consist of | | Description | OS | | — | — | — | | random | 32GB Ram 16 vCPU - Master, Compute, Infra | Fedora 29 Workstation | | banana | 32GB Ram 8 vCPU - Compute | Fedora 29 Server | | randomstore | 8GB (16gb When i find some RAM) RAM 4 vCPU, Compute | Fedora 29 Server |

This is NOT an HA environment, as much as I would like an HA environment I cannot get away with any more machines being on 24/7

I have got SSH Keys working between the environments and correctly set up the hostnames.

I have cloned the ansible playbook repository to /opt/src and checked out 3.11

This is my initial inventory file that I have put into /etc/ansible/hosts

# Create an OSEv3 group that contains the masters, nodes, and etcd groups
[OSEv3:children]
masters
nodes
etcd
glusterfs

# Set variables common for all OSEv3 hosts
[OSEv3:vars]
# SSH user, this user should allow ssh based auth without requiring a password
ansible_ssh_user=root
# If ansible_ssh_user is not root, ansible_become must be set to true
#ansible_become=true
openshift_deployment_type=origin

# uncomment the following to enable htpasswd authentication; defaults to AllowAllPasswordIdentityProvider
#openshift_master_identity_providers=[{'name': 'htpasswd_auth', 'login': 'true', 'challenge': 'true', 'kind': 'HTPasswdPasswordIdentityProvider'}]

# host group for masters
[masters]
random.phillips11.cf

# host group for etcd
[etcd]
random.phillips11.cf

# host group for nodes, includes region info
[nodes]
random.phillips11.cf OpenShift_node_group_name='node-config-all-in-one' ansible_python_interpreter='/usr/bin/python3' OpenShift_schedulable=True
randomstore.phillips11.cf OpenShift_node_group_name='node-config-compute' OpenShift_schedulable=True ansible_python_interpreter=/usr/bin/python3
banana.phillips11.cf OpenShift_node_group_name='node-config-compute' OpenShift_schedulable=True ansible_python_interpreter=/usr/bin/python3

[nodes:vars]
ansible_python_interpreter='/usr/bin/python3'

[glusterfs]
random.phillips11.cf glusterfs_devices='[ "/dev/fedora/gfs" ]'
banana.phillips11.cf glusterfs_devices='[ "/dev/fedora_banana/gfs" ]'
randomstore.phillips11.cf glusterfs_devices='[ "/dev/fedora/gfs" ]'

The video of the PreReq running is available here.

asciicast