Know when your website is down — before your customers tell you
OwlHound checks your site every 5 minutes from outside your network and emails you the moment an outage is confirmed. No false alarms, no passwords, nothing to install.
What we check
You give us the address of any public page — your homepage, a checkout page, a booking form, a /health endpoint. Every 5 minutes we request it the way a visitor's browser would. A page counts as down when it times out, the domain stops resolving, the connection is refused or reset, the TLS certificate is invalid, or the server answers with an error such as 500, 502, 503 or 404.
If a site answers with 401, 403 or 429 it is clearly alive but protecting itself from automated requests (a login wall, a firewall, rate limiting), so we treat it as reachable rather than waking you up for nothing.
Alerts you can actually trust
An alert that cries wolf gets ignored, so we spend our effort on being right:
- Retries inside every check. A failed request is retried twice more over the next 30 seconds before it counts.
- A second confirming round. One failed round is only a suspicion. The outage is confirmed — and you are emailed — when the next round fails too.
- We check ourselves first. Before reporting a failure we confirm our own server can reach the internet and resolve DNS. If our side is the problem, nothing is sent.
- One email per outage. You get a single "down" message with the cause (for example HTTP 503 or timed out after 15s) and a single "recovered" message with the total downtime. Rapid flapping is rate-limited.
- Email and Slack or Discord. Every alert can also post to a webhook channel.
Planned work without the noise
Deploying a new version or moving hosts? Schedule a maintenance window for one site or all of them. We keep checking, but outages during the window are not emailed and are not counted against your uptime.
A weekly report you can read in a minute
Every Monday you receive a clean HTML report: uptime percentage per site, each outage with its cause and duration, average response time, a 7-day availability strip, and how many days remain on every SSL certificate and domain registration. Agency plans send it under your own brand.
What we don't do (yet)
We run HTTP(S) checks from our own servers, so a routing problem between us and your host could, in rare cases, look like an outage — our confirmation steps make that unlikely but not impossible. We only monitor public URLs: we can't log in to a site, check a private network, ping a port or monitor a cron job.
Frequently asked questions
How often do you check my website?
Every 5 minutes. A confirmed outage is therefore reported within roughly 5 to 10 minutes of starting, because a failure must be repeated in a second round before we email you.
What counts as downtime?
Timeouts, DNS failures, refused or reset connections, invalid TLS certificates, and HTTP 5xx, 404 or 410 responses. Pages that answer with 401, 403 or 429 are treated as reachable.
Will I get false alarms?
We confirm every failure with retries and a second round, and check our own network before alerting. That removes almost all false alarms, but no external monitor can promise zero.
Can I monitor more than my homepage?
Yes. Any public URL works — a checkout page, a booking form, a pricing page or an API health endpoint.
Do I need to install anything?
No. Paste a URL and we start checking. We never ask for passwords or access to your servers.
Start watching your site in minutes
7-day trial. No installs and no passwords — we only check public pages.