Skip to content

Mainly the inventory files I used to install openshift and if any prep work needed on atomic hosts.

Notifications You must be signed in to change notification settings

kempy007/atomic-openshift-install

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 

Repository files navigation

atomic-openshift-install

This repo is mainly storing the inventory files I used to install openshift and notes if any prep work needed on atomic hosts.

The objective is to use a container from within the target cluster to bootstrap the installation of an openshift cluster by leveraging docker.io/openshift/openshift-ansible particular attention needs to be paid to the versions as they need to match the tags being pulled.

Host setup

download https://cloud.centos.org/centos/7/atomic/images/CentOS-Atomic-Host-7.1808-Installer.iso and install. can use packer to automate this section but the kickstart file needs to be different than the regular OS kickstart (do manual install and pull off kickstart file). you could then parametize the kickstart and feed back into an ansible gui job that fires off packer tasks.

edit network config

edit /etc/sysconfig/network-scripts/ifcfg-enp0s3 enable onboot

expand root volume so can download container

lvdisplay check the lv path ignore beginning /dev/ as middle section is hostname or cah lvextend -r -L 20GB cah/root this command expands root volume to 20GB

setup ssh keys between all hosts, including self from primary node(one you decided to use, I use a master).

ssh-keygen ssh-copyid root@ip.ip.ip.ip

run ansible container to install cluster

So we need to match the version in the inventory file otherwise cluster install will fail. atomic install --system --storage=ostree --set INVENTORY_FILE=/root/inventory --set PLAYBOOK_FILE=/usr/share/ansible/openshift-ansible/playbooks/prerequisites.yml --set OPTS="-v" docker.io/openshift/origin-ansible:vINT.INT.INT

atomic install --system --storage=ostree --set INVENTORY_FILE=/root/inventory --set PLAYBOOK_FILE=/usr/share/ansible/openshift-ansible/playbooks/deploy_cluster.yml --set OPTS="-v" docker.io/openshift/origin-ansible:vINT.INT.INT

to start fresh and remove whole cluster

atomic install --system --storage=ostree --set INVENTORY_FILE=/root/inventory --set PLAYBOOK_FILE=/usr/share/ansible/openshift-ansible/playbooks/adhoc/uninstall.yml --set OPTS="-v" docker.io/openshift/origin-ansible:vINT.INT.INT

References

Finally redhat are talking about all-in-one installs at https://blog.openshift.com/openshift-all-in-one-aio-for-labs-and-fun/ but most importantly it has another inventory template referenced at https://blog.openshift.com/openshift-all-in-one-aio-for-labs-and-fun/

About

Mainly the inventory files I used to install openshift and if any prep work needed on atomic hosts.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published