n8n-nodes-vivere

The Vivere node for n8n. It tells you when a workflow stopped running, finished without doing anything, or failed.

n8n's Error Workflow only fires when an execution fails. A Schedule Trigger that never fires (the instance was restarting at that minute, the workflow got deactivated, the timezone is not the one you think) leaves no execution behind, so nothing reports it. Vivere expects a ping on a schedule and alerts you when one does not arrive in time.

Setup

1. Install the node. In n8n, open Settings → Community Nodes → Install and enter n8n-nodes-vivere.
2. Add your API key. Create a key at vivere.dev/app/keys, then in n8n add a Vivere API credential and paste it in.
3. Ping at the end of the workflow. Add a Vivere node after the last step, choose Ping → Success, and pick the monitor from the list. Set the monitor's period to the trigger's schedule, so a workflow on Every hour gets period 1h. If you have no monitor yet, create one in Vivere or with the node's Monitor → Create operation.

That is enough to be alerted when the workflow stops running. The rest of this page is optional.

Operations

Ping

Every ping operation sends one ping per execution, however many items reach the node, and passes its input items through unchanged, so the node can sit in the middle of a workflow.

• Start marks the start of a run, so Vivere records how long it took. Put it right after the trigger.
• Success reports a finished run. Options:
  • Item Count: how many items the run handled, shown on the run in Vivere. Items Received (the default) counts the items that reached the node, Custom takes a number or an expression, Don't Report sends none. A single empty item, which is what a node with Always Output Data passes on when it found nothing, counts as 0.
  • Fail If 0 Items: when the count is 0, report the run as failed instead. A run that finished without handling anything then alerts like a broken one.
  • Usage: model, input tokens, output tokens, cache read and cache write tokens, and cost in USD. Map them from an AI node's output with expressions, and Vivere shows what each run cost and can alert when a monitor's cost passes a limit.
  • Run Log: text to keep with the run, such as a summary.
• Fail reports a failed run, which marks the monitor down at once instead of waiting for the deadline. Message is optional. Left empty, the node sends the error it finds in its input: from an Error Trigger that is the error message, the node that failed, the workflow name and a link to the execution; from a node's error output it is the error message.

The monitor can be picked from the list, pasted as its ping URL, or given as its ID.

Monitor

• Create makes a heartbeat monitor with a name, a schedule (a period such as 1h, or a cron expression with a timezone) and a grace, and returns it, including its ping_url. Optional: the page to put it on, tags, a description, and whether it is public.

Grace is how late a ping may be before you hear about it. Allow for the run's longest normal duration plus the delay before n8n starts it; the grace calculator works it out.

Example workflow

examples/nightly-sync.json is a daily workflow with all three pings. Import it with Import from File in the workflow menu, then pick your monitor in the three Vivere nodes.

• Run started pings Start after the Schedule Trigger.
• Fetch rows has On Error set to Continue (using error output) and Always Output Data on.
• Run finished pings Success with Fail If 0 Items on, so a day that fetched nothing alerts.
• Run failed sits on the error output and pings Fail with the error message.

Using it from an Error Trigger

If your workflows already share an Error Workflow, add a Vivere node after its Error Trigger with Ping → Fail and leave Message empty. The run in Vivere then shows the error, the node that failed, the workflow and a link to the execution.

With one monitor per workflow, set Monitor to ID and pick the ID by workflow name with an expression:

{{ ({ 'Nightly sync': '0f6c3a52-8d1e-4b7a-9c2f-5e4d3b2a1c0f', 'Invoice export': '7d2e9b10-4c3a-4f5e-8a6b-1c2d3e4f5a6b' })[$json.workflow.name] }}

Keep the Success ping in each workflow as well. The Error Trigger covers runs that fail; only a missing Success ping covers runs that never happen.

Credentials

The Vivere API credential takes an API key from vivere.dev/app/keys (keys start with vv_). n8n checks it by listing your pages. The node uses the key to list monitors and pages and to create monitors. The pings themselves go to the monitor's ping URL and do not carry the key.

Compatibility

Tested on n8n 2.43.7. The package has no runtime dependencies.

Resources

• Vivere's n8n guide
• Why an n8n Schedule Trigger stops firing
• Vivere API reference
• n8n community nodes documentation

License

MIT