Deploy a specific hyperkube imageΒΆ

By default fuel-ccp-installer uses an hyperkube image downloaded from the quay.io images repository. See the variables hyperkube_image_repo and hyperkube_image_tag variables in the kargo_default_common.yaml file.

To use a specific version of hyperkube the hyperkube_image_repo and hyperkube_image_tag variables can be set in the deploy-k8s.sh script. This is done through the CUSTOM_YAML environment variable. Here is an example:

#!/bin/bash
set -ex

# CHANGE ADMIN_IP AND SLAVE_IPS TO MATCH YOUR ENVIRONMENT
export ADMIN_IP="10.90.0.2"
export SLAVE_IPS="10.90.0.2 10.90.0.3 10.90.0.4"
export DEPLOY_METHOD="kargo"
export WORKSPACE="${HOME}/workspace"
export CUSTOM_YAML='hyperkube_image_repo: "gcr.io/google_containers/hyperkube-amd64"
hyperkube_image_tag: "v1.3.7"
'

mkdir -p $WORKSPACE
cd ./fuel-ccp-installer
bash -x "./utils/jenkins/run_k8s_deploy_test.sh"

In this example the CUSTOM_YAML variable includes the definitions of the hyperkube_image_repo and hyperkube_image_tag variables, defining what hyperkube image to use and what repository to get the image from.

Note

If you use an inventory Git repo please refer Inventory and deployment data management to know how you can set variables for the environment.

Previous topic

Diagnostic info collection tools

Next topic

Configurable Parameters in Fuel CCP Installer

Project Source

This Page