Internal health checks pass while the site is unreachable from the outside. Cronitorex calls your endpoints from its own infrastructure, the way a user would, and tells you when reality and your dashboards disagree.
HTTP 200 with an empty body is still an outage. Each check defines what a healthy response looks like, and anything else counts as a failure.
GET https://api.example.com/health interval 60s / 30s assert status 200 · body =~ "ok" timeout 5s last result 200 in 187 ms p99 (24h) 312 ms
Networks flicker. When a check fails with a transient error, Cronitorex retries with backoff: after 60 seconds, then 5 minutes, then 15 minutes. If a retry passes, no alert is sent.
Hard failures skip the ladder. When your endpoint returns 500 or refuses connections outright, the alert goes out immediately, and a recovery notice follows once checks turn green again.
Checks run every 60 seconds on Free and Pro, and every 30 seconds on Business. Worst case detection is one interval plus the first retry, not whenever someone happens to open the site.
Every check also records response time, so you see the slow week before the down day. Trends live on the same dashboard as your cron jobs and certificates.
Checks every 60 seconds, retries with backoff, response time history on every check.
Checks every 30 seconds, for endpoints where each minute of downtime has a price tag.
From Cronitorex infrastructure in the EU, outside your network. A check that runs next to your app misses the failures your users actually hit, like DNS, TLS or routing problems.
No. Retries are internal. You get one alert when a failure is confirmed and one recovery notice when checks pass again, on email, Telegram, Discord, Slack or a webhook.
Yes. Add custom headers such as an Authorization token, pick the HTTP method and attach a request body. Header values are stored encrypted and never included in alert messages.
Sign up, get an API key, paste the curl into your cron. That's it.