Running a deployment and reading run history

Once a workflow is deployed in CloudRadial AutomationAI, you can run it and review every execution in History. This is the final setup step, run and verify: trigger a run, watch it execute end to end, and read the run log to confirm the whole loop works before you point real work at it. This article covers running a deployment, the statuses and trigger types a run can have, how a run's outcome is decided, and drilling into a run's steps. It is for all roles.

In this article:

• Running a deployment
• The History view
• Run statuses
• How a run's outcome is decided
• Trigger types
• Drilling into run steps
• Cancelling a run

Running a Deployment

A run executes the deployment's pinned published version. You start one with the Run action in the Designer, choosing the target deployment to launch. The work is queued rather than run on the spot: the deployment's runner picks it up on its next poll and reports progress back, so a run begins as pending and advances as the runner works. Runs also start on their own from a schedule or a webhook trigger.

The History View

Open History (/runs) to see every workflow execution across your workspace, newest first. Each row shows the workflow, status, trigger, the user who started it, the runner, start and completion times, duration, and step count. On wider screens, an On finish column shows the result of the workflow's On finish lane beside the run status, or a dash when no lane ran. The view updates live as runs progress, so a run appears and its status changes without a manual refresh. You can sort and filter the columns and page through the history. Opening a run from the Runners dashboard or from a workflow scopes the view to that runner or workflow.

Run Statuses

A run carries one status as it moves through its lifecycle:

• Pending, queued, waiting for the runner to pick it up
• Running, the runner is executing the run's steps
• Finishing, the main path has ended and the workflow's On finish lane is reporting the outcome; the run is still active
• Succeeded, every step on the main path completed successfully
• Failed, a step errored, its branch stopped at that step, and the run did not complete
• Timed out, a step exceeded its allowed time, or the run exceeded the workspace run ceiling, and the run was stopped
• Cancelled, someone stopped the run before it finished

For what to do when a run fails or times out, see Troubleshooting a failed or timed-out run.

How a Run's Outcome Is Decided

A step that fails or times out ends its branch of the workflow: the steps connected after it do not run. Steps on other parallel branches that are already under way can still finish.

The run's outcome is set by the first step on the main path that failed or timed out, and nothing that happens afterward changes it. A run where one branch fails and another branch later succeeds still ends Failed. A run whose only failing step timed out ends Timed out. If no main-path step fails or times out, the run ends Succeeded. A For Each node follows its failure policy: with continue-on-error, failed items do not stop the steps after the loop; with fail-fast, the first failed item ends the branch.

If the workflow has an On finish lane connected for that outcome, the run shows Finishing while the lane runs, then ends with its main outcome. The lane's own result is recorded separately as the On finish result and never changes the run status, so a run whose work succeeded but whose ticket note failed still shows Succeeded, with an On finish result of Failed. If the outcome's output is not connected, or the workflow has no On finish node, the run ends without entering Finishing. For how to build a lane, see Reporting a run's outcome with the On finish lane.

Trigger Types

The Trigger column shows how each run started:

• Scheduled, fired by a Routine on its cron cadence
• Manual, started by a person with the Run action
• Webhook, started by an inbound call to the workflow's webhook trigger

Drilling Into Run Steps

Click a run to open its detail. Beside the run's Status, the detail shows a Main path field with the main path's outcome while the run is Finishing, and an On finish field with the lane's result once a lane has run. A Summary section gives a plain-text account of the run: a status line, who requested it when that is known, one line per main-path step (up to the 50 most recent) with its status, duration, and result or error, and a link to the run. It is the same text a workflow step can read as {{ run.summary }}.

Below the summary, the detail lists the run's main-path steps in order, followed by any On finish lane steps, each marked On finish. Each step shows its language, its status, and its sections: the step's Output, its Log if one was captured, and its Error if it failed. The first step is often the synthetic trigger step, which shows the trigger payload, the manual or webhook JSON the run started with, rather than a runner result. An open run detail updates live as the runner reports each step.

Cancelling a Run

A run that is going nowhere does not have to be waited out. Select Cancel on the run’s row in History, or the Cancel action in the run detail panel, and the run is finalized as cancelled. Any steps still outstanding are closed out with it, and a result the runner reports afterwards is discarded. Cancelling a run that is Finishing stops only its On finish lane: the lane's result is recorded as cancelled, and the run ends with its main outcome. The same action is available on a workflow’s own run history and on an agent run started from the AI Playground.

Your workspace also has a run ceiling, a maximum wall-clock time any single run may stay open. A run that passes it is finalized as timed out with the reason recorded on the run, so a run waiting indefinitely on an agent step cannot sit open forever. The ceiling does not end a run that is Finishing; the On finish lane has its own, shorter time budget. The ceiling is generous by default and exists as a backstop, not a step budget; for step-level limits see Troubleshooting a failed or timed-out run.

If you are still having trouble, we're here to help! Submit a ticket here for assistance, and don't forget to check our status page to ensure there are no outages in your area.