fagan2888 / cloud-build-demo

Code for a Google Cloud Build demo run at the SLC Python meetup

Google Cloud Build Demo - SLC Python Meetup

Slides for this talk

Code

Code for a Google Cloud Build demo run at the SLC Python meetup. To experiment with these builds, create a GCP project and modify the cloudbuild.yaml files to point to your project. For example, in

steps:
- name: 'gcr.io/cloud-builders/docker'
  entrypoint: 'docker'
  args: [ 'build', '-t', 'gcr.io/slc-python-demo/airflow', '.' ]
images: ['gcr.io/slc-python-demo/airflow']

change slc-python-demo to your project id.

The easiest place to run these builds is in the Google Cloud Shell. There are four builds: airflow-install-and-test, build-airflow-container, test-from-airflow-container and pull-and-write-data. Change to the appropriate directory and run

gcloud builds submit

The file dag.py in two of the builds is taken from a standard tutorial in the Apache Airflow Project.

airflow-install-and-test

Pulls a Python 3.7 container from Docker Hub, installs Apache Airflow and runs a basic test.

build-airflow-container

Builds an Airflow container by pulling a Python 3.7 container and pip installing Airflow. Pushes the resulting container to the GCP project container registry.

test-from-airflow-container

Pulls the Airflow container built in build-airflow-container and runs a simple test.

pull-and-write-data

Not a true build - downloads AWS IP ranges, stores the data to a file, and pushes the file to a Cloud Storage bucket. You should create a target bucket in your project and update pull-data.sh.