Sleep through the night.
Uptime checks your sites and APIs on schedule, tells you only when a failure is confirmed, and never calls a missing check healthy.
100% free, open source alternative to UptimeRobot, Pingdom and Better Stack
Free and open source, MIT licensed.
Honest evidence
Green means it was checked.
Every green dot is a check that actually ran and passed. When a result never arrives or expires, Uptime marks that time Unknown instead of guessing, so a quiet runner can't pass for a healthy site.
Unknownno check received
02:4502:5002:5503:0003:0503:1003:1503:2003:2503:30Confirmed incidents
Woken only by real failures.
You choose how many failed checks in a row open an incident and how many passes close it. Each incident keeps the evidence that justified it: the status, timing and the rule that failed on every check.
3 failures in a row → incident opened
2 passes → recovered 07:15
06:45200 OK06:50200 OK06:5550307:0050307:05missing "ok"07:10200 OK07:15200 OKChecks
Check what matters.
HTTP and HTTPS checks can assert on status, body text, headers, JSON values and response time. Keyword checks look for the words a page should contain, and heartbeats notice when a cron job or backup stops reporting in.
GEThttps://api.example.com/health
- Status is 200
- JSON status equals "ok"
- Responds within 800 ms
Nightly backup · heartbeat received 02:00
Notifications
Told where you already look.
Send incidents and recoveries by email, to a Slack incoming webhook, or to a signed JSON webhook. Every enabled destination covers every monitor in your installation.
Down: Checkout API
3 failed checks in a row · 503 Service Unavailable
#ops
Signed JSON webhook
{
"state": "recovered",
"title": "Recovered: Checkout API",
"incidentKey": "…"
}Ownership
Yours, start to finish.
Run Uptime on your own computer. A small private runner checks your targets from a machine you choose, and you can export monitors, evidence and delivery history whenever you like.
- Your own computer
- Your runner, your network
- Export any time
$git clone https:// github.com/ just-every/ run-uptime.git uptime$cd uptime$rer install$rer setup$rer dev
Also included
The quiet details.
Maintenance windows
Pause alerting while you work on a service.
Manual checks
Run a check now without opening or closing an incident.
Heartbeat tokens
Each heartbeat uses its own token and increasing sequence numbers.
Pinned connections
Every DNS answer and redirect is validated, then the connection is pinned.
Signed runner results
The Worker accepts only results signed by your runner.
Safe restores
Restored monitors start paused and imported alerts never send.
Rest easy tonight.
Open-source monitoring you run yourself.
Free and open source, MIT licensed.