Skip to content
English
Do you need help? Customer portal
neowit.io
  • There are no suggestions because the search field is empty.

Job failed

When a specific workflow execution shows up as an error in the Executions tab, here's how to find out why and what to do next. For workflow owners and anyone else debugging a red run.

Open the failed run

  1. Open the workflow in the editor and switch the toolbar mode from Editor to Executions, or expand the Executions strip on the left.
  2. Optionally filter the list by error at the top.
  3. Click the red row. The canvas overlays that run's state and the Logs panel fills in with the task list and the failed node's details.

Heads up: execution URLs don't include a job id today, so you can't share a direct link to a specific run. When reporting to support or a teammate, include both the workflow id and the execution id from the row (the first eight characters shown are enough).

Read the error

The Logs panel's TASKS pane lists every node the run touched. Click the task with the red icon. The OUTPUT pane shows the error message. The INPUT pane shows the data that flowed into that task after templates were resolved — invaluable when you need to know whether the problem was the config or the incoming data.

Full walkthrough of the panel in Testing your workflow.

[SCREENSHOT] The Logs panel on a failed execution. The TASKS pane on the left shows five tasks with the fourth one red. The INPUT pane shows the document that entered that task; the OUTPUT pane on the right shows the error message that stopped the run.

Common causes

Integration disconnected — "Not authenticated"

The action's integration is in Bad credentials or Unable to connect state. Reconnect it — see Integration disconnected.

Data reference or template error

Any of these strings usually points at a template:

  • "Config resolution failed: …"
  • "template syntax error: …"
  • "value not found"
  • "cannot convert template result … to …"

Open the node's config, click into the failing field's Data reference picker or Template editor, and verify the path actually exists on the INPUT pane for that run. Full mechanics in References and templates.

Node configuration

  • "If contains no conditions" / "invalid operation specified" / "invalid regex: …" — the If node is misconfigured. Open it in the editor, fix the condition rows, save.
  • "no dispatch handler for task (specId=…)" — the node's type no longer exists (e.g. its integration was deleted, or the node type was retired). Remove the offending node and replace it.

Timeout — "Timed out"

A task took longer than its configured timeout. Most core nodes default to one minute; integration actions vary. If the task genuinely needs more time, check whether the node has a timeout setting in its config. If a run gets stuck for days rather than minutes, the Stale job timeout (Workflow Engine settings, default 2 days) auto-errors it — that usually points at an infrastructure glitch, not a workflow bug.

Stop And Fail — "Stop And Fail Reached" (or your custom message)

Not a bug. A Stop And Fail node ran on purpose, typically on a condition's false branch. If the run shouldn't have reached it, fix the condition logic upstream.

Re-run a failed job

There is no retry button on past executions. The Run from here and Test this node actions on the canvas apply to the current in-session execution, not to old history.

To replay the failed event manually:

  1. Open the trigger node in the editor.
  2. Click Import from execution in the trigger's input dialog.
  3. Pick the failed run. Its trigger payload lands in the JSON editor.
  4. Click Run workflow.

For recurring or future events, once you've fixed the underlying issue the next real trigger runs against the corrected workflow automatically. The red failed run stays in history but doesn't block anything.

Stop a run that's going wrong

If a run is in flight and you can see it's going to fail (or spam Slack, or toggle lights you don't want toggled), click Stop execution on the toolbar. The job is marked canceled. In-flight external calls on the integration side may still complete — stop is not a time machine.

When to contact support

Most failures have a clear error message in OUTPUT and fit one of the buckets above. Reach out with the workflow id and execution id if you see:

  • A job with no error message that went stale after days.
  • "no dispatch handler" that persists after you've re-saved the workflow.
  • An error you can't identify from this article.

Related