startrhacm

Deploy Red Hat Advanced Cluster Management (RHACM) via ClusterPool

./startrhacm.sh

Features

• Claims an OpenShift cluster from a ClusterPool and deploys Upstream or Downstream RHACM of any available branch (x.x), version (x.x.x), or snapshot
• (optional) Automatically resizes a specified pool if the pool isn't large enough
• (optional) Patch the cluster to accept connections from localhost paths for development
• startrhacm.sh can be aliased and run from anywhere on your computer:

  alias startrhacm="${PATH_TO_STARTRHACM}/startrhacm.sh"

Prerequisites

Clone the following repos:

• Lifeguard - Collection of scripts to claim from ClusterPools
• Deploy - Installation scripts for RHACM
  • Be sure to set up your pull secret for Deploy (see Deploy documentation)
• Pipeline - Collection of available RHACM snapshots (private repo)
  • Exception: Pipeline is not necessary if the full snapshot is provided via RHACM_SNAPSHOT

Setup config scripts

Exports are contained in a config.sh script (but they can also be exported outside of the script if desired). To set up config.sh for your own use, customize utils/config.sh.template or your squad-specific template below as desired and rename it to utils/config.sh.

• Required: Set paths to cloned repos

  export LIFEGUARD_PATH= # Path to local Lifeguard repo
  export RHACM_DEPLOY_PATH= # Path to local Deploy repo
  export RHACM_PIPELINE_PATH= # Path to local Pipeline repo (optional only if deploying downstream or RHACM_SNAPSHOT is specified directly)

• Optionally configure other variables as indicated in the comments in utils/config.sh.template (if optional variables are not provided, the startrhacm.sh script will prompt you for ClusterClaim options and will deploy the latest upstream RHACM snapshot)
• Set up file permissions for the config script to be executable

  chmod +x ./utils/config.sh

Configure oc CLI to point to the Collective cluster

You'll need to be logged in to the Collective cluster that hosts ClusterPools. If you're not, startrhacm.sh will detect this and provide you with the link to the login command. (If you're using a different ClusterPool cluster, you can disable this check by setting export DISABLE_CLUSTER_CHECK="true")

Squad-specific config.sh Templates

• GRC - config.sh.template-grc

Extras

• Shrink and/or expand ALL ClusterPool sizes on a schedule using a CronJob. By default, schedules are set to shrink to 1 at 8 PM EST (1 AM UTC) every day and expand to 2 at 6 AM EST (11 AM UTC) Monday - Friday

  cd extras
  export CLUSTERPOOL_TARGET_NAMESPACE=<namespace>
  export SERVICE_ACCOUNT_NAME=<service-account-name>
  ./clusterpool-shrink.yaml.template.sh
  ./clusterpool-expand.yaml.template.sh
  oc apply -f .