Know when Grafana needs you, without living in dashboards.
I run my own Grafana, and most days I need one answer: is anything on fire? Not the cathedral of panels I built at 1am while pretending this was a normal hobby. Just yes or no.
It is a homelab, not a $100k incident room with someone saying “customer impact” into a headset, but it is still my stuff. I want to know before my wife asks why Netflix is not working, which is how this house files support tickets.
The alert apps I tried wanted dashboards, phone-app cosplay, or Prometheus algebra. Beacon does the opposite: stays quiet, then gets impossible to miss when something actually needs me.
It immediately tells you whether things are all clear, warning, critical, or unreachable. That last one matters, because a monitor that is dead but still smiling at you is not monitoring, it is interior decoration.
Paste your Grafana URL and a token. That is the setup. No label-matcher warm-up act, no configuration pilgrimage, no pretending you enjoy onboarding.
Silence the alert, open its dashboard, or jump to the runbook right from the notification. Fix it immediately, or begin the respected engineering tradition of staring at it first.
A monitor that nags gets muted, then becomes expensive wallpaper. Beacon keeps quiet until there is something worth interrupting you for.
For a homelab or small infrastructure, see what mine is watching →
If you run a serious on-call operation with escalation policies, paging schedules, and a room full of people saying “customer impact,” use a serious on-call tool. If you live inside Prometheus label matchers across twelve backends, Beacon will feel too simple by lunch. It is for one person, or a small Mac-heavy team, watching their own Grafana without turning monitoring into a second job.
