Point Hesklo at a URL and it requests it on your interval. Match an exact status code or accept any 2xx, and optionally require a string in the response body so a page that loads but returns nothing still counts as down.
A server can answer with a 200 and still be broken. HTTP monitoring lets you say what a healthy response actually looks like.
Require a specific code. Useful when an endpoint is meant to return a 201 or a 301 and anything else means something changed.
The forgiving option. Anything in the success range passes, so a redirect chain or a varying code does not page you.
Require a string to appear in the response. Catches the case where the app serves a shell page with an empty or errored payload.
Every check records how long the request took, so you can see a service degrading before it fails outright.
Any HTTPS monitor tracks days remaining on the certificate. Branch on it in a policy and get warned ahead of time.
Set how long to wait before a request counts as failed, so a slow endpoint is judged on its own terms.
Require a number of consecutive failed checks before the monitor is marked down. One dropped request never wakes anyone.
Probe the URL on demand and see the exact response, without waiting for the next scheduled run.
Pause the monitor during planned work. It keeps its history and resumes exactly as it was.
A cooldown on each notify step, so a service bouncing up and down does not turn into a hundred messages.
Three monitors are free forever. Add a check, draw what happens when it fails, and watch the policy fire.