Tutorial: Project run-failure alert automation

Build a run-failure alert that sends a Slack notification when a run in your project fails.

This tutorial walks you through building a project automation triggered by run status: when a run in your project transitions to Failed, W&B sends a Slack notification.

For guidance creating a Registry automation, see Tutorial: Registry artifact alias automation.

Prerequisites

• A W&B project.
• A Slack integration configured in Team Settings.

Create a project automation

Follow these instructions to set up a project-scoped automation: When a run in the project transitions to Failed, W&B sends a Slack notification.

W&B App

1. Open the project and click the Automations tab in the sidebar, then click Create automation.
2. Choose the event Run state change. Set the state to Failed. Optionally add a run name or user filter to limit which runs trigger the automation.
3. Click Next step. Set Action type to Slack notification and select the Slack channel.
4. Click Next step. Give the automation a name (for example, “Run failure alert”) and an optional description, then click Create automation.

For more detail, see Create a Slack automation (Project tab).

Python

Use the public API to create the same automation with a project scope and run-state filter:

import wandb
from wandb.automations import OnRunState, RunEvent, SendNotification

api = wandb.Api()

project = api.project("my-project", entity="my-team")
slack_integration = next(api.slack_integrations(entity="my-team"))

event = OnRunState(
    scope=project,
    filter=RunEvent.state.in_(["failed"]),
)
action = SendNotification.from_integration(slack_integration)

automation = api.create_automation(
    event >> action,
    name="run-failure-alert",
    description="Notify the team when a run fails.",
)

Test the automation

Create a run and log it to the project, explicitly marking it as failed:

import wandb

with wandb.init(project="my-project") as run:
    run.log({"loss": 1.23})
    run.finish(exit_code=1)

Within a short time you should see a Slack message with the run link and status.

Run the full tutorial in a notebook

Run the Automations tutorial notebook to create and test project-scoped and registry-scoped automations.

Download or clone the repo and open the notebook in Jupyter, Google Colab, or your preferred notebook environment.

Go further

• Automation events and scopes for all project and registry event types.
• Create a Slack automation and Create a webhook automation for full UI and payload details.
• Manage automations with the API for list, get, update, and delete examples.
• Automations API reference for all event and action classes.