Catch n8n Workflows That Fail Silently with the Doyouok Node
n8n is great at running jobs at 2 a.m. while you sleep. It is less great at telling you when one of those jobs quietly did nothing.
Two things go wrong with scheduled workflows. Sometimes the workflow doesn't run at all: the server was down, the workflow was deactivated, or a node failed and stopped the run. And sometimes it runs perfectly and still gets it wrong: the CRM sync synced 0 contacts, the import processed 0 rows, the API answered without success. n8n shows that second kind as a green, successful execution, so nobody notices for days.
The Doyouok node for n8n catches both. It is free, open source and takes one node at the end of your workflow.
Set up in three steps
- Install the node. On self-hosted n8n, open Settings > Community Nodes, press Install and enter
n8n-nodes-doyouok. - Connect your account. Sign in to Doyouok (the Free plan is enough), open Profile > API tokens and press Create token. In n8n, create a Doyouok API credential and paste the token.
- Add the node at the end of any workflow, after the steps that do the work. Save the workflow and run it once.
That's it. There is no ping URL to copy and no check to create by hand. With Check set to This Workflow (Automatic), the node creates a check named after your workflow on the first run and keeps it in sync when you rename the workflow or change its schedule.
Tell it what success looks like
- Expected Every and Unit: how often the workflow should run, for example 1 Days for a nightly job. If no run arrives in time, you get an overdue alert.
- Result: Check Conditions reports success only if the data reaching the node matches your conditions. Success and Failure report a fixed result, handy on an error branch.
- Conditions: any field from the previous step, for example
{{ $json.syncedContacts }}is greater than 0.
So a nightly sync that brings in 120 contacts is reported as healthy. One that runs but syncs 0 contacts is reported as failed, and you hear about it the same morning instead of next week.
Alerts that say what went wrong
Each alert names the workflow, links straight to the failed execution in your n8n, and says why it failed: either your own Failure Message (for example "Synced {{ $json.syncedContacts }} contacts") or the condition that wasn't met, such as "syncedContacts is 0, expected > 0". Alerts go to email, and to Telegram, Slack or Discord if you connected them in Doyouok.
Want to know the moment a step errors instead of waiting for the overdue alert? Set the step's On Error to Continue (using error output) and connect its error output to a second Doyouok node with Result set to Failure. Both nodes report to the same check.
Only the result leaves your n8n
The node sends the workflow's name and ID, the expected interval, whether the run succeeded and, on a failure, the reason and a link to the execution. Your workflow's input and output data stay in n8n. The node is MIT licensed and the source is on GitHub.
Free for your first 5 workflows
The node itself is free. Each monitored workflow uses one check from your Doyouok plan, and the Free plan includes 5 checks with no credit card. Bigger plans are on the pricing page.
Questions about setup, conditions or alerts? The Doyouok for n8n help page has a step-by-step guide, an FAQ and an assistant that answers on the spot.