Edit Vertex AI Workbench instance's external IP.

Hi,

Recently I have migrated my Jupyter Notebook from Vertex AI’s USER-MANAGED NOTEBOOKS instance to Workbench Instance. I want to edit the external IP of the underlying VM on which Workbench Instance is running but I am getting permission issues:

“Required ‘Current principal doesn’t have permission to mutate this resource!’ permission for ‘wb-instance’”

As per this documentation, I can not edit using Console or Compute Engine API and only way to edit is using gcloud workbench instance update command. But I could not find anyway to update the external IP using gcloud workbench command.

https://cloud.google.com/vertex-ai/docs/general/troubleshooting-workbench#unable_to_edit_underlying_virtual_machine

Can anyone please tell if there is anything I am missing?

1 Like

To update the external IP of a Workbench instance using the gcloud command, you would typically use the gcloud compute instances update command, as Workbench instances are based on Compute Engine VMs. However, since Workbench is a managed service, direct manipulation of underlying Compute Engine resources might be restricted due to permissions and service agreements.

If the documentation states that you should use gcloud workbench instance update, it’s possible that the functionality to update the external IP address directly isn’t available at the moment. In such cases, it might be best to reach out to Google Cloud Support for further assistance or clarification on the available options for updating the external IP address of a Workbench instance.

Here’s what you can do:

  1. Check Documentation Updates: Make sure to periodically check the Google Cloud documentation for updates or changes in the supported commands and features.

  2. Reach Out to Support: If you’re facing permission issues or if the functionality you need isn’t documented, it’s best to contact Google Cloud Support. They can provide guidance specific to your situation and might be able to assist you with updating the external IP address or suggest alternative solutions.

  3. Consider Alternatives: While direct manipulation of the underlying Compute Engine resources might not be possible due to service agreements, there might be alternative solutions or workarounds available. Google Cloud Support can provide insights into these options based on your requirements and constraints.

1 Like