I’ve been working on Quorum, an open-source tool that monitors n8n workflows from the outside. It catches cases where a workflow stops running, keeps failing, or finishes successfully with zero useful items.
Existing workflows can be monitored through polling, so you don’t have to modify them. Quorum opens an incident when something goes wrong, keeps the history in one place, and resolves it automatically when the workflow recovers.
It’s self-hosted, Docker-based, has no telemetry, and is currently in beta.
I’m looking for a few people running real n8n workflows who want an easier way to notice problems before a client or customer does! I’ll help connect the first workflow and configure the expected schedule and alerts.
https://github.com/werniq/quorum_core
How do you currently notice when an important workflow silently stops doing its job?