Uptime monitoring
How often your site is checked, what counts as down, the Availability card, the address to watch, and downtime alerts.
Every proved domain is also checked to see whether its site answers, between audits. If it stops answering, you get an alert, then another one when it is back, with how long it was down.
How often
- every 5 minutes on Free;
- every 3 minutes on Solo;
- every minute on Agency and Agency Plus;
- every minute on every plan while a site is down.
What counts as down
A check fails when the site gives no answer within 10 seconds, when its name does not resolve, when the connection is refused, when its certificate is not valid, or when the server answers with an error (5xx). A redirect, a login page or even a 404 counts as an answer: the server is up.
A site is declared down after two failed checks in a row. The first answer after that ends the incident.
A site that has never answered since monitoring started is shown as Not reachable, without any alert: it may simply live at another address. Set the address to watch (below) and alerts start once it has answered once.
The Availability card
On the domain’s page, the Availability card shows:
- whether the site is up or down, and since when;
- its uptime over the last 24 hours, 7 days and 30 days;
- its response time hour by hour over the last 24 hours, with failed checks;
- the latest incidents, with their cause and duration.
In the domain list, a domain whose site is down is marked down.
The address to watch
By default, the home page of the domain over https is checked. In the card’s Availability settings, owners and admins can:
- set another address to watch, such as
https://www.example.com/or a health page, on the domain or one of its subdomains, over https on the standard port; - untick Check that the site answers to turn the checks off for this domain.
Downtime alerts
Downtime and recovery alerts go to the channels of the Alerts page: by e-mail on every plan, and by webhook from the Solo plan. Each channel can refuse them: press No downtime alerts on its line (and Downtime alerts to take them again). The rule of a channel (changes, issues, always) applies to audit alerts, not to downtime.