Every maintenance plan we sell includes the same promise: we notice before your customers do. Sentinel is how we keep it.
The brief
The monitoring tools we tried had one of two problems. Either they watched one thing well (uptime, but not certificates; response codes, but not performance drift), or they watched everything and buried the signal in dashboards nobody opened twice. We wanted the whole picture for every property in our care, and we wanted it quiet: a tool that says nothing until something needs saying.
So we built it, the way we'd build it for a client.
The build
Sentinel is a Laravel application with a Vue front end, running checks around the clock from multiple regions. It watches the things that actually take sites down:
- Uptime and response health, checked frequently enough that an incident is measured in minutes, not mornings.
- TLS certificates, because expiry is the most preventable outage on the internet.
- DNS history, so a wrong or hijacked record is caught as a change, not discovered as a mystery.
- Performance over time, because slowness is usually the first chapter of downtime.
Incidents open and resolve themselves as conditions change, alerts reach a human immediately, and status pages let clients see what we see.
Where it stands
Sentinel runs the standing watch behind every Rootstuff maintenance plan, and it's available as a product in its own right. It's also the reason this studio talks about uptime with a straight face: the 99.9% record on our homepage is Sentinel's measurement, not our memory.
Building and operating our own production SaaS keeps the studio honest. Every opinion we have about deployments, monitoring, and maintenance gets tested on our own product first.