New Scriptworker-Script

This procedure is based on the experience of adding the pushmsixscript to scriptworker-scripts.

• Write your new script and add it to scriptworker-scripts.

• You may want to clone-and-modify an existing script.

• Consult the documentation.

• Write unit tests for your script; you can run them locally with tox.

• Add clients to ci-configuration repo

• Example

• Add a new yaml file for your script’s secrets to the sops repo.

• You may want to clone-and-modify an existing file.

• ed25519PrivateKey is the same for all scripts.

• Populate taskclusterAccessToken by resetting the access token for each client in the taskcluster UI: open each of your clients, use the … menu in lower right, select “reset access token”, then scroll up to see the alert.

• Add your configuration parameters to cloudops-infra, like this.

• Reference those parameters in your script’s init_worker.sh and worker.yml. Example.

• Ask SRE to create a repo on hub.docker.com for your new script.

• Try pushing to your script’s dev branch:

• git push origin master:dev-<your-script>

• Add entries to k8s-autoscale like this.

• The docs may be helpful.

• Merging will automatically trigger a dev deploy.

• Be sure to deploy k8s-autoscale to production as well.

• When you are ready to start running tasks with your new script, add configuration to taskcluster, like this.