jnpacker / acm-hive-provision-yamls

acm-hive-provision-yamls

Working with this repository

Edit cluster-templates

1. First decide which providers you want to use. If you want to use all three providers, then skip the next step
2. Delete the provider-install-config.yaml and cluster-provider.yaml files for providers you don't want to use. Example:

rm cluster-templates/azure-install-config.yaml
rm cluster-templates/cluster-azure.yaml

Also remove the deleted file references from ./kustomization.yaml 3. Edit the cluster-templates/PROVIDER-install-config.yaml file

• Add a value for ssh public key, to be used to connect to the cluster nodes
• For Azure you must also add a value for baseDomainResourceGroupName

4. Edit the cluster-templates/cluster-PROVIDER.yaml file for each provider

• Add value for pull-secret, ssh-private-key and creds (all of these are secrets)
• The creds value(s) are different for each provider type

5. Values in the files you edited for 3 and 4 can be customized as you wish to template cluster deployments for your environments
6. Update the *_BASEDOMAIN variables in the ./deploy.sh script

Deploying clusters

1. Connect to the OpenShift cluster running Red Hat Advanced Cluster Management for Kubernetes
2. Run ./deploy.sh NAME_PREFIX VERSION where NAME_PREFIX is combined with the PROVIDER type to make a name and VERSION is the OpenShift version that will be deployed. If you Hub cluster does not have a ClusterImageSet for the version you choose, the script will fail and provider you with a list of available versions.
3. Run oc apply -k . to provision the clusters. If you want this step to happen automatically, uncomment the last line in the ./deploy.sh script
4. Alternately, you can modify the fully formed Cluster YAML's before you apply them