Skip to content

Latest commit

 

History

History
33 lines (20 loc) · 1.33 KB

README.md

File metadata and controls

33 lines (20 loc) · 1.33 KB
Raw

Infrastructure

Tools and packages that are used to run a deployment of Taskcluster.

Kubernetes

Using

Before using this, you need a python 3 installation with a couple packages installed via pip install -r requirements.txt

To generate a helm chart for all the services, run ./bin/helmit.py. To limit it to a specific service, use the --service flag followed by the service name.

The chart is written to the chart directory. You will need to set values for all the variables in values.yaml in order to apply it.

The generated ingress currently assumes you are deploying to GCP.

Explanation

The structure is

  • templates/ - The json-e templates
  • services/ - The service-specific context for evaluating the json-e templates
  • ingress/ - charts for ingresses you can choose from
  • bin/ - Tools to generate helm from json-e

A file in services/ is YAML and has three or four sections:

  1. Project name - used for identification
  2. Secrets - a map of keys and vlues for the service secrets
  3. Deployments (optional) - a list of maps with keys and values used for configuring the deployment
  4. Cronjobs (optional) - a list of maps with keys and values used for configuring the cronjob

A value in the YAML can be either a literal string or it can be a placeholder for a value helm will substitute. The latter all start with .Values..