Public IP for OCI Data Science Notebook Sessions

Question

Pulumi · Accepted Answer

To set up a Data Science Notebook Session on Oracle Cloud Infrastructure (OCI) with a Public IP, we would generally follow these steps:

1. Create a Virtual Cloud Network (VCN) and a subnet within OCI that will host the Data Science Notebook Session.
2. Create the Data Science Notebook Session within the subnet.
3. Associate a Public IP address with the notebook session (if the subnet is public or if using a private subnet, we would need to set up a NAT Gateway to provide internet access).

Here's a basic Pulumi program in Python that sets up the necessary resources in OCI to host a Data Science Notebook Session with a Public IP:

```python
import pulumi
import pulumi_oci as oci

# Configuration: Replace these with appropriate values.
compartment_id = 'ocid1.compartment.oc1..xxxxxx'
subnet_id = 'ocid1.subnet.oc1..xxxxxx'

# Create a new Public IP resource to be associated with the notebook session
public_ip = oci.core.PublicIp("notebookPublicIp",
                              compartment_id=compartment_id,
                              lifetime="RESERVED",  # Can be either RESERVED or EPHEMERAL
                              )

# Create the Data Science Notebook Session. This session will utilize the previously created Public IP.
notebook_session = oci.data_science.NotebookSession("notebookSession",
                                                    compartment_id=compartment_id,
                                                    project_id="ocid1.datascienceproject.oc1..xxxxxx",  # Update this with your project's OCID
                                                    notebook_session_configuration_details=oci.data_science.models.NotebookSessionConfigurationDetailsArgs(
                                                        shape="VM.Standard2.1",  # Choose an appropriate VM shape
                                                        subnet_id=subnet_id,  # Use the VCN subnet that has been set up for notebook sessions
                                                        # If the subnet is private, we would need to associate the NatGateway here
                                                    ),
                                                    notebook_session_runtime_config_details=oci.data_science.models.NotebookSessionRuntimeConfigDetailsArgs(
                                                        custom_environment_variables={},  # Optional: Set any environment variables needed for the notebook
                                                    ),
                                                    display_name="MyDataScienceNotebookSession",
                                                    )

# Export the public IP address and the notebook session URL for easy access
pulumi.export('notebook_session_url', notebook_session.notebook_session_url)
pulumi.export('public_ip', public_ip.ip_address)
```

This program configures a Data Science Notebook Session in OCI and assigns it a Public IP Address. The variable `compartment_id` should be set to the identifier of your OCI compartment where the resources will reside. Similarly, `subnet_id` should be set to the OCID of the VCN subnet that will be hosting the notebook session.

The `oci.core.PublicIp` resource is used to create a Public IP that can be associated with a service like a Data Science Notebook Session to access it publicly over the internet.

Then, we create a `oci.data_science.NotebookSession` specifying the required configurations like the shape (size of the VM), `subnet_id` where the VM will be located, and an optional display name.

Finally, the details of the created resources (the public IP and the notebook session URL) are exported, which will then be displayed as output when you run `pulumi up`.

Please replace the placeholder values (like `ocid1.compartment.oc1..xxxxxx`) with your actual OCI identifiers in order to use this Pulumi code in your environment. Ensure that your environment is appropriately configured with the required OCI credentials before running Pulumi commands.