Host Gatus without installing anything
A public status page where every probe is written down in a file: HTTP, TCP, DNS or ICMP, with conditions spelled out, right down to the certificate expiry date.
Starting at €2.99/month
Introduction
Gatus installed, updated and backed up by us, on a machine in France. A plain, fast public status page that shows your users your services are running. You install nothing and configure no server: the application is online a few minutes after you order, under your own name.
The plan that suits it is App Découverte at €2.99/month incl. VAT — 1 GB of memory. A probe hitting an address once a minute costs next to nothing: Gatus is a Go binary that keeps its results and renders a static page. That gigabyte covers the checks running in parallel, the history kept to draw the curves, and the system underneath. There is no reason to pay more for monitoring that must, by design, stay light.
One point deserves to be said plainly: a status page is only useful if it does not go down together with what it watches. Installing it on the server it observes is the classic mistake — the day that machine stops, the page meant to announce it stops too. Here it runs on our side, on infrastructure separate from yours.
What Gatus can do
Conditions written down, not guessed
A probe passes if what you wrote is true: status code equal to 200, response time under 300 ms, the JSON body's "status" field equal to "healthy", certificate valid for more than another 48 hours. A service can answer 200 and still be broken; Gatus lets you say so.
HTTP, TCP, DNS, ICMP and TLS
Web pages are not the only thing you can check: opening an SMTP port, resolving a DNS record, sending a ping, watching the expiry of a certificate or of a domain name are probes like any other.
Alerts into what you already use
Email, Slack, Discord, Telegram, Mattermost, ntfy, Gotify, PagerDuty, Opsgenie or a plain webhook. You set how many consecutive failures trigger an alert, and the recovery message goes out on its own.
A public page with history
Availability over 24 hours, 7 days and 30 days, a response-time curve, services grouped into families. It is a page you hand to your customers, not a dashboard reserved for the team.
What people actually use it for
Answering before anyone asks
An agency publishes a status page for its clients. When an application drags, the client checks the page instead of phoning, and the team does not spend the morning saying the same thing ten times over.
Watching the checkout, not the home page
An online shop probes the payment page and checks that the certificate protecting it stays valid. The home page can render perfectly well while payment no longer goes through.
Checking an API is healthy, not merely alive
A vendor queries its health endpoint and requires the response body to state that the database is reachable. A service answering "OK" without its database is a service that is down.
Watching the invisible infrastructure
A mail server on its ports, a DNS record that must point to the right place, a domain name approaching expiry: the longest outages often start there, because nobody is looking.
Not being paged during a planned upgrade
A maintenance window declared in advance avoids the 3 a.m. wake-up for planned work. The probes carry on, the alerts stay quiet.
Hosted in France
In our own infrastructure in Nouvelle-Aquitaine, which we operate ourselves — that is, somewhere other than the machine you are watching.
Frequently asked questions
Nothing. You choose the application, fill in the fields it asks for — often an email address and a name — and it goes online. Access arrives by email with the address and the credentials.
The way you configure it. Gatus is described in a YAML file: the list of probes can be reread, compared from one version to the next and restored identically, which suits teams that keep their configuration under version control. Uptime Kuma is driven entirely with the mouse, which is quicker when you add a probe now and then. Both are in the catalogue; the choice is about how you work, not about how good the monitoring is.
You need to know what it measures. Gatus tells you whether your service is reachable — or not — from our network, in France. That is exactly what you need to catch an outage, an expired certificate or a service answering badly. It is not a worldwide probe network: if your problem is comparing response times as seen from Tokyo and São Paulo, this is not the tool.
Three settings cover it: a number of consecutive failures before the alert fires, which absorbs brief blips; a maintenance window when work is planned; and the automatic recovery message, so you are not left with an open alert on an incident that is already over.
Yes, they are part of the service. A self-hosted application left un-updated is a security risk, not a saving.