This serverless deployment will create a Cloud Function and a sample Cloud Scheduler recipe. Only one StarThinker Cloud Function is required to run all StarThinker Recipes. The Cloud Scheduler defines the task, credentials, and schedule for each recipe.
This only needs to be done once, after cloud functon creation, all recipes are run via scheduler configuration. This deployment leverages the latest StarThinker PYPI package, which is the last tagged version in the repository.
bash install/deploy.sh
- Set up Google Cloud Project.
- Option 4) Deploy Cloud Function
- Note the trigger URL and start using the scheduler to run recipes.
- A sample job is is created on cloud scheduler.
- It will run a hello world task.
- Extend the sample with other recipe tasks.
- When extending recipes, each {"field":{...}} entry in a recipe must be replaced with actual values.
- Service cerdentials are only used if a task auth is set to service.
- User credentials are only used if a task auth is set to user.
- Client credentials are only used if a task auth is set to user, and user credentials are empty.
- Online utility to make scheduler easier to use.
Next, review list of available tasks in the Recipe Gallery, view Scripts, or check Command Line Helpers.
- Google Cloud Functions - where your function will deploy.
- Google Cloud Scheduler - where your recipe workflows will deploy.
- Google Cloud StackDriver Logs - where your logs will be written.
- Google Cloud Credentials - where you manage your credentials.
- Google Cloud Billing - examine costs in real time.
© 2021 Google LLC - Apache License, Version 2.0