Scripts and tools for automating Kayobe operations. Intended for use in CI/CD pipelines.

kayobe-automation is typically installed as a Git submodule within a Kayobe configuration repository. This allows the version of kayobe-automation in use to be controlled.

Check out your kayobe-config repository and add the submodule.

cd <kayobe-config-repository>
git submodule add https://github.com/stackhpc/kayobe-automation.git .automation
cd .automation
git checkout 0.0.1
cd ..
git add .automation
git add .gitmodules
git commit -m "Add kayobe-automation"

Where 0.0.1 is the tag of the release you want to use. The submodule will act like a separate git repository, so if you make any changes you will need to change into .automation directory and commit those separately.

You can change the branch at a later date with:

cd .automation
git checkout <new-tag>
cd ..
git add .automation
git commit -m "Update kayobe-automation"

Configuration for kayobe-automation is typically provided via files in a .automation.conf directory at the root of the kayobe-config repository. The following files are supported:

• config.sh: Mandatory bash script that will be sourced by kayobe-automation scripts. Typically it will export environment variables necessary to configure kayobe-automation.

Use the examples in examples/* for inspiration.

KAYOBE_URI: Deprecated Kayobe source repository URI. It is preferred to install Kayobe via a requirements.txt file in kayobe-config.

KAYOBE_BRANCH: Deprecated Kayobe source repository branch name. It is preferred to install Kayobe via a requirements.txt file in kayobe-config.

KAYOBE_ENVIRONMENT: Optional Kayobe environment name.

KAYOBE_AUTOMATION_CONFIG_PATH: Path to kayobe-automation configuration directory. Defaults to .automation.conf in the kayobe-config root directory.

KAYOBE_AUTOMATION_SSH_PRIVATE_KEY: Private key used to login to kayobe-managed hosts

KAYOBE_AUTOMATION_SSH_PRIVATE_KEY_NAME Name of private key used to login to kayobe-managed hosts. Defaults to id_rsa.

KAYOBE_AUTOMATION_LOG_LEVEL: Verbosity of logging. Choose one of: debug, info, warn, error

KAYOBE_VAULT_PASSWORD: Kayobe vault password.

KAYOBE_AUTOMATION_PR_AUTH_TOKEN: (Required when KAYOBE_AUTOMATION_PR_TYPE is set) Auth token to use when submitting pull-requests. Typically you need to create a service account and generate a personal access token.

TEMPEST_OPENRC: OpenRC file content to use when running Tempest

kayobe-automation provides various pipeline scripts in the pipeline directory. Many of these are wrappers around Kayobe commands.

kayobe-automation provides a Dockerfile to build a container image for running commands in docker/kayobe/.

Building the image requires the use of Docker buildkit.

On Ubuntu, you may need to do the following:

sudo apt install golang-docker-credential-helpers

To build, from the root of the kayobe-config repository:

DOCKER_BUILDKIT=1 docker build --file .automation/docker/kayobe/Dockerfile --tag kayobe:latest .

To use the image to run one of the pipeline scripts, from the root of the kayobe-config repository:

docker run -it --rm -v $(pwd):/stack/kayobe-config kayobe:latest /stack/kayobe-config/.automation/pipeline/overcloud-host-configure.sh

It may be necessary to provide certain environment variables at runtime, e.g. KAYOBE_ENVIRONMENT or KAYOBE_VAULT_PASSWORD. This may be done via docker run -e ENV_VAR=value ....

You need to customise /etc/gitlab-runner/config.toml to add support for docker in docker:

[[runners]]
  name = "seed"
  url = "https://gitlab.com/"
  token = "redacted"
  executor = "docker"
  [runners.custom_build_dir]
  [runners.cache]
    [runners.cache.s3]
    [runners.cache.gcs]
    [runners.cache.azure]
  [runners.docker]
    tls_verify = false
    image = "centos:8"
    privileged = true
    disable_entrypoint_overwrite = false
    oom_kill_disable = false
    disable_cache = false
    volumes = ["/certs/client", "/cache", "/opt/kayobe/images"]
    shm_size = 0

This also uses a volume so that IPA images can be shared between pipelines.

Dependencies: go

Install the git hooks:

[stack@seed .automation]$ python3 -m venv ~/will/venv/pre-commit
[stack@seed .automation]$ source ~/will/venv/pre-commit/bin/activate
(pre-commit) [stack@seed .automation]$ pip install pre-commit
(pre-commit) [stack@seed .automation]$ pre-commit install

This will run the formatter on commit.

To run manually:

(pre-commit) [stack@seed .automation]$ pre-commit run --all-files