Monitor Silent n8n Workflows and Alert on Stalled Workflows

What This Workflow Does

This workflow acts as a self‑monitoring health check for your n8n instance. It detects active workflows that have not run within a configurable time window — a failure mode that produces no failed executions and therefore does not trigger standard error alerts. By running daily, it catches silently stalled workflows so you can investigate before they become a problem.

The workflow uses the n8n REST API to fetch all active workflows and their recent executions, then compares the last execution time against a threshold (default: 26 hours). Any workflow that is active but has not run within that window is flagged as “silent”. You can attach your own notification node (email, Slack, etc.) at the end to alert your team.

Node‑by‑Node Breakdown

• Every morning — Schedule Trigger: Runs the workflow daily at 8:00 AM (configurable in the node’s rule). No authentication needed.
• Configuration — Set node: Defines two variables used by later nodes:
  • baseUrl: Your n8n instance address (example: https://n8n.example.com). Edit this value.
  • staleHours: Number of hours a workflow can be silent before it is considered stalled (default: 26).
• Get workflows — HTTP Request: Fetches all workflows from the n8n API (GET /api/v1/workflows?limit=250). Uses Header Auth (credential type httpHeaderAuth with key X-N8N-API-KEY). Requires an n8n API key.
• Get executions — HTTP Request: Fetches the most recent 250 executions (GET /api/v1/executions?limit=250). Also uses Header Auth with the same credential. Both requests share the same API key credential for simplicity.
• Find silent workflows — Code node (JavaScript): Processes the two API responses. It:
  • Filters active (non‑archived) workflows.
  • Builds a map of the latest execution timestamp per workflow.
  • Compares each active workflow’s last run against the current time minus staleHours.
  • Outputs an array of objects with workflow name, workflowId, lastRun, hoursSilent, and a note.
  • If no silent workflows are found, outputs a summary: { silent: 0, activeChecked: <count> }.

Setup Instructions

1. Create an n8n API key in your n8n instance: Settings → n8n API → Create an API key. Copy the generated key.
2. Add a Header Auth credential in n8n:
   • Name: e.g., n8n API Key
   • Type: Header Auth
   • Header Name: X-N8N-API-KEY
   • Header Value: Paste the API key from step 1.
3. Configure the Configuration node:
   • Set baseUrl to the URL of your n8n instance (no trailing slash), e.g., https://my-n8n.example.com.
   • Set staleHours to the number of hours after which a workflow is considered stalled (recommended: slightly longer than your expected run interval).
4. Add an alert node after Find silent workflows (e.g., Slack, email, webhook) to notify you when silent workflows are detected.
5. Activate the workflow. It will run daily and check all active workflows.

Use Cases and Adaptations

• Core use case: Catch workflows that have stopped running due to schedule misconfiguration, resource exhaustion, or API failures that don’t produce error executions.
• Alerting integration: Replace the final Code node with an IF node that branches to your preferred notification service (Slack, email, Telegram, PagerDuty) when silent workflows are found.
• Custom thresholds: Adjust staleHours per workflow type. For example, a once‑daily report can tolerate 28 hours, but a 5‑minute webhook should be checked after 10 minutes.
• Run more frequently: Change the Schedule Trigger to run every hour for tighter monitoring.
• External monitoring: For maximum reliability, run this workflow from a separate n8n instance (or external system) to avoid the blind spot described in the sticky note.

{
  "name": "n8n Silent Workflow Check",
  "id": "duskwatchSilentCheck",
  "nodes": [
    {
      "parameters": {
        "content": "## Silent workflow check\n\nFinds workflows that are **active but have not run** — the failure mode that produces no failed execution and therefore no error workflow alert.\n\n### Setup (2 minutes)\n1. In n8n: **Settings → n8n API → Create an API key**.\n2. On **Get workflows**, create a *Header Auth* credential:\n   - Name: `X-N8N-API-KEY`\n   - Value: your API key\n   Reuse the same credential on **Get executions**.\n3. In **Configuration**, set `baseUrl` to your n8n address (no trailing slash) and `staleHours` to how long silence is still normal.\n4. Attach your own alert node after **Find silent workflows**.\n\n### The catch\nThis workflow lives inside the instance it watches. If that instance is down, restarting, or its\nschedule triggers did not re-register, this check does not run either — and its silence looks\nexactly like everything being fine.\n\nThat is the one failure it cannot cover. Watching from outside is the only thing that does:\nhttps://duskwatch.me/checklist?ref=template",
        "height": 620,
        "width": 460
      },
      "id": "8f1c0a10-0000-4000-8000-000000000001",
      "name": "Read me",
      "type": "n8n-nodes-base.stickyNote",
      "typeVersion": 1,
      "position": [
        -460,
        -120
      ]
    },
    {
      "parameters": {
        "rule": {
          "interval": [
            {
              "triggerAtHour": 8
            }
          ]
        }
      },
      "id": "8f1c0a10-0000-4000-8000-000000000002",
      "name": "Every morning",
      "type": "n8n-nodes-base.scheduleTrigger",
      "typeVersion": 1.2,
      "position": [
        40,
        100
      ]
    },
    {
      "parameters": {
        "assignments": {
          "assignments": [
            {
              "id": "a1",
              "name": "baseUrl",
              "value": "https://n8n.example.com",
              "type": "string"
            },
            {
              "id": "a2",
// ... truncated (copy to see full JSON)