SiteBrief/Documentation

Uptime Monitoring

SiteBrief checks your site every 1–5 minutes, 24/7. The moment it goes down you get an alert — and another when it recovers.

How it works

At each check interval, SiteBrief makes an HTTP or HTTPS request to your site's URL from our primary monitoring server. We measure the response time and check the HTTP status code. If the site returns a non-success status code or fails to respond within the timeout window, we flag it as down and send alerts.

When a site goes down, we continue checking. As soon as it responds normally again we mark it upand send a recovery alert.

StatusMeaningWhat triggers it
🟢 UpSite is responding normallyHTTP 200–399 within timeout
🔴 DownSite is not respondingHTTP 400+, timeout, connection refused, DNS failure
🟡 DegradedResponding slowlyResponse time exceeds threshold set in settings

What counts as 'down'

A site is marked down when any of the following happen:

  • HTTP 4xx / 5xx — the server responded but with an error code (404, 500, 503, etc.)
  • Connection timeout — the server didn't respond within 30 seconds
  • Connection refused — the port is closed or the server actively rejected the connection
  • DNS resolution failure — the domain doesn't resolve to any IP address
  • SSL/TLS error — the certificate is invalid, expired, or the handshake failed
  • Keyword not found — if keyword monitoring is enabled and the text is missing from the response
ℹ️
Note:HTTP 301 and 302 redirects are followed automatically (up to 5 hops). If the final destination returns 200, the site is marked as up.

Check intervals

PlanIntervalChecks per day per site
FreeEvery 5 minutes288
StarterEvery 5 minutes288
AgencyEvery 1 minute1,440
Agency ProEvery 30 seconds2,880
💡
Tip:If your client's site is an e-commerce store or SaaS product where every minute of downtime means lost revenue, upgrade to Agency or Agency Pro for 1-minute or 30-second checks.

Uptime percentage

The uptime percentage shown in the dashboard is calculated over the last 7 days by default. You can view 30-day and 90-day uptime in the site detail charts.

Formula: (total checks − failed checks) / total checks × 100

A single failed check in a 5-minute interval window counts as 5 minutes of downtime. For example, 1 failed check out of 288 daily checks = 99.65% uptime for that day.

Response time tracking

Every successful check records the time from sending the HTTP request to receiving the first byte of the response. This is shown in:

  • The Recent checks table on the site detail page
  • The Response time chart (30-day history with 7/30/90 day selector)

You can set a response time threshold in settings. If a check exceeds this threshold (e.g. 3000 ms), the site status changes to degraded and an alert is sent.

Dual confirmation

When dual confirmation is enabled, SiteBrief requires two consecutive failed checks before sending a down alert. This eliminates false positives caused by brief network blips between our server and your site.

  • Check 1 fails → SiteBrief immediately re-checks
  • Check 2 also fails → alert is sent
  • Check 2 succeeds → no alert, site is considered up
⚠️
Warning:Dual confirmation adds one check interval of delay before alerting. On a 5-minute interval, you may not get alerted until 10 minutes after the site went down. Disable it on Agency plans where checks run every minute.

HTTP method

By default, SiteBrief makes a GET request. You can change this to HEAD or POSTin the site settings.

MethodWhen to use
GET (default)Most websites — downloads the full page to check content and keywords
HEADWhen you only want to check status codes and headers, not download the body (faster, but keyword monitoring won't work)
POSTAPI endpoints that require a POST request — set a JSON or form body in the Request body field

Environments

Each site can be tagged as Production, Staging, or Development. This is for your organisation only — it affects how the site is labeled in the dashboard and in client reports. Monitoring behavior is identical across environments.

Pausing monitoring

You can pause monitoring for a site until a specific date and time. While paused, no checks run and no alerts are sent. Use this during planned maintenance windows or deployments.

💡
Tip:Set a pause before you start a maintenance window, not after. This prevents spurious down alerts during the work.

Frequently asked questions

My site is up but SiteBrief says it's down — what's happening?
The most common cause is a firewall or CDN rate-limiting our monitoring IPs. Check if your site returns 429 or 403 to unknown user agents. You can add SiteBrief's user agent to your allowlist: SiteBrief-Monitor/1.0. Also check if your host blocks requests from the EU (where our primary server is based).
I got an alert but the site was only down for 10 seconds — is that normal?
Yes. If the check happened to run exactly when the site was down, we flag it. With dual confirmation enabled, you need two consecutive failures, so very brief blips are filtered out.
Does SiteBrief follow HTTPS redirects from HTTP?
Yes. Redirects are followed up to 5 hops. If your site redirects from http:// to https://, we follow it and check the final HTTPS URL.
Will I get alerted for every failed check, or just the first one?
Just the first one. Once a site is marked down, we suppress further down alerts until it recovers. You'll then get one recovery alert.
Can I monitor a page behind a login?
Not directly via the UI. A workaround is to create a public health-check endpoint (e.g. /health) that returns 200 and monitor that instead. For API endpoints requiring auth, you can use the keyword or JSON assertion feature with a POST request and an auth token in the request body.
What is the check timeout?
SiteBrief waits 30 seconds for a response. If no response is received within 30 seconds, the check fails with a timeout error.