Triggers GCP Cloud SQL export backups to GCS.

Solution documentation:

https://cloud.google.com/solutions/scheduling-cloud-sql-database-exports-using-cloud-scheduler

• main.py - the code
• requirements.txt - the pip modules to bootstrap
• deploy.sh - upload the code and deps

Upload the function to GCF in the current GCP project - this script will call gcloud functions deploy with the required switches:

./deploy.sh

• a Cloud PubSub topic
• Cloud Scheduler jobs to trigger backups
  • see gcp_cloud_schedule_sql_exports.sh in DevOps Bash tools repo
• a service account with permissions to access Cloud SQL
  • see gcp_sql_create_readonly_service_account.sh in DevOps Bash tools repo
• each Cloud SQL instance to be backed up requires objectCreator permissions to the GCS bucket
  • see gcp_sql_grant_instances_gcs_object_creator.sh in DevOps Bash tools repo

Instead of deploy.sh you can alternatively use the Serverless framework for which a serverless.yml config is provided:

serverless deploy

If this is your first time using Serverless then you'll need to install the GCP plugin:

serverless plugin install --name serverless-google-cloudfunctions

The serverless.yml config expects to find $GOOGLE_PROJECT_ID and $GOOGLE_REGION environment variables.

Serverless requires additional permissions for the service account: Deployment Manager Editor and Storage Admin to create deployments and staging buckets.

You can also build a serverless artifact to .serverless/ without deploying it (generates Google Deployment Manager templates and a zip file - useful to check what would be uploaded / ignored):

serverless package