Skip to content

Lab: Initial Volume Mounts

Before the Kubeflow Pipeline run starts, Rok takes a snapshot of the workspace and data volume. Then, it clones these snapshots to create new volumes that the pipeline steps will use. The initial clones of the workspace volume and data volume are mounted to each of the containers that are created to process the Kubeflow pipeline steps.

Requirements

You can identify where these volumes are mounted by reviewing the pipeline steps.

Review the pipeline step clean_data to identify where the volumes are mounted.

Solution

View Solution

The volumes will be mounted to the /home/jovyan directory, with sub directories created for the data volumes.