Skip to main content

Metadata ingestion setup for Google Cloud Storage

You must grant the Model-Prime ingest role read access to the GCS bucket in which your robologs are stored in order for the platform to process them.

This guide requires no prerequisite GCS knowledge. If you already know how to configure GCS, or do so using cloudformation/terraform/gsutils, you may skip to the Grant Model-Prime Permission to the Bucket section, which will provide you with the IAM information to use with your existing workflow.

Step 1. Create a bucket, if necessary​

If you already have a bucket which contains your robolog data, skip this step.

Create a GCS bucket that will store robologs to be ingested by Model-Prime. You may do so by following the official GCS bucket creation user guide.

Feel free to select any Location when creating the bucket.

Step 2. Grant Model-Prime permission to the bucket​

Model-Prime needs the storage.objectViewer permission on your bucket to ingest the data it contains. This permission may be granted within the bucket's Permissions tab as shown below.

Bucket_Permissions

Once in this tab, scroll down to the Permissions section and press Grant Access.

Grant_Access

Once editing, add the Storage Legacy Bucket Reader and Storage Object Viewer roles to the bucket for the Model-Prime Principal, which can be found in our web application.

Add_Role

Next steps​

Model-Prime will now have the ability to read objects from your bucket. In order to receive information about what exists in the bucket, ingestion must be invoked. Proceed to the next section to learn more about the options available for invoking ingest.