Configure email notifications

Vertex AI Pipelines can notify you of the success or failure of a pipeline run. When the pipeline exits, Google Cloud sends a final status notification email to the email addresses that you specify.

This guide shows how to configure email notifications from a pipeline by using the Email notification component in the Google Cloud Pipeline Components SDK.

Before you begin

Before you build a pipeline that sends notifications, use the following instructions to set up your Google Cloud project and development environment.

  1. To get your Google Cloud project ready to run ML pipelines, follow the instructions in the guide to configuring your Google Cloud project.

  2. Install v1.8 or later of the Kubeflow Pipelines SDK.
    (Optional) Before installing, run the following command to see which version of the Kubeflow Pipelines SDK is currently installed:

      pip freeze | grep kfp
  3. To use Vertex AI Python client in your pipelines, install the Vertex AI client libraries v1.7 or later.

  4. To use Vertex AI services in your pipelines, install the Google Cloud Pipeline Components.

Send a notification from a pipeline

The following example shows how to configure email notifications by defining an email notification task (notify_email_task) and adding it to the pipeline's exit handler (dsl.ExitHandler). This notification task invokes the VertexNotificationEmailOp operator in the email notification component when the pipeline exits.

from kfp import dsl
from kfp import compiler
from google_cloud_pipeline_components.v1.vertex_notification_email import VertexNotificationEmailOp

def TASK_NAME():
    notify_email_task = VertexNotificationEmailOp(recipients=RECIPIENTS_LIST)

    with dsl.ExitHandler(notify_email_task):
        # Add your pipeline tasks here.


Replace the following:

  • PIPELINE_NAME: The name of the pipeline.

  • PIPELINE_ROOT_PATH: Specify a Cloud Storage URI that your pipelines service account can access. The artifacts of your pipeline runs are stored within the pipeline root.

    The pipeline root can be set as an argument of the @kfp.dsl.pipeline annotation on the pipeline function, or it can be set when you call create_run_from_job_spec to create a pipeline run.

  • TASK_NAME: The name of the pipeline task for which you're configuring email notifications.

  • RECIPIENTS_LIST: A comma-separated list of up to three email addresses to send the notification email to.

Add your pipeline tasks in the body of the dsl.ExitHandler exit handler function. By wrapping the tasks in the exit handler function in this way, you specify that the notification email component should report the status of these tasks when the pipeline exits. For example, if a task within the exit handler's contents fails, then the notification reports the status as failed.

Example of a notification email

If you configure email notifications for a pipeline using the code sample in Send a notification from a pipeline, Vertex AI sends an email notification similar to the following when the pipeline exits:

Subject: Vertex Pipelines job "PIPELINE_NAME" task "TASK_NAME"
From: Google Notifications <>

Hello Vertex AI Customer,

Vertex Pipelines job "PIPELINE_NAME" task "TASK_NAME" ended with the following state: {status}.

Additional details:
- Project: {project}
- Pipeline name: PIPELINE_NAME
- Pipeline job ID: {pipeline_job_id}
- Start time: {start_time}

To view this pipeline job in Cloud Console, use the following link: {console_link}

The Google Cloud AI Team

In this example:

  • {status} represents the final state of the task, which can be SUCCEEDED, FAILED, or CANCELLED.
  • {project} is the project name.
  • {pipeline_job_id} is the unique pipeline job ID.
  • {start_time} represents the start time for the pipeline.
  • {console_link} is a hyperlink to the pipeline job in the Google Cloud console.