Cloudera Docs

Stopping a workflow

You can stop a running workflow to safely terminate AI agents and halt active processes. This provides essential control over your environment, allowing you to interrupt long-running tasks, adjust inputs, or conserve resources.

Stopping a workflow is useful in the following scenarios:

  • Unexpected duration: The workflow is taking significantly longer than anticipated.
  • Input adjustments: When changes are required for the input data before the process completes.
  • Unsatisfactory results: The current output is not meeting your requirements or expectations.
  • Process redundancy: The workflow is no longer needed to complete its task.
  • Fresh start: To restart the process with entirely different settings.

The Stop Workflow feature ensures that the workflow terminates cleanly. All progress is saved, and compute resources are properly released.

Stopping a workflow

  1. In the Cloudera console, click the Cloudera AI tile.

    The Cloudera AI Workbenches page displays.

  2. Click on the name of the workbench.

    The workbench Home page displays.

  3. In the left navigation pane, click the AI Studios option.
  4. Click the Launch button in the Agent Studio box to initiate installation in runtime mode. The Configure Studio: Agent Studio page is displayed.
  5. Set the environment variables for the Agent Studio.
  6. Select the Editor as PBJ Workbench.
  7. Under Runtime, select Agent Studio as the Kernel
  8. Click Launch AI Studio.

    The AI Studio Setup Steps page is displayed.

    After launching, you can view the list of tasks being executed as part of the AI studio deployment.

  9. After configuration, Agent Studio is displayed in the left navigation page under AI Studios.
  10. Click Agent Studio and click Get Started to orchestrate AI agents to collaborate on complex tasks, powered by custom tools and seamless workflow automation.
  11. Create your workflow. You can either stop the workflow during Testing, in the Tools Playground or stop the workflow after Deploying it.
  12. To terminate the workflow, click the red Stop button.
  13. The button changes to display Stopping while the request processes. The workflow ends, and the button disappears once the process is complete.

The workflow typically terminates within seconds. Once stopped:

  • Resource cleanup is handled automatically.
  • You can immediately initiate a new workflow or modify existing settings.