Monitors
Understand how HTTP, ping, and port monitors work, including check intervals, prober regions, and degraded state.
What is a monitor?
A monitor is an active probe that the platform runs on a schedule. At each interval, a prober node sends a request to your endpoint and records the result: status, response time, and any errors. If the result does not match your configured expectations, a failure is counted.
A typical monitor's 90-day record looks like this — each segment is one day of check history. Hover any day to see the exact uptime percentage.
Production API · sample monitor
99.94%Monitor types
| Type | Protocol | Typical use |
|---|---|---|
| HTTP / HTTPS | HTTP 1.1 / 2 | Web services, REST APIs, health endpoints |
| Ping | ICMP | Network reachability of any host or IP |
| Port (TCP) | TCP | Databases, SMTP, custom services |
Check interval
The check interval determines how often the prober sends a request. Shorter intervals detect outages faster but consume more checks from your plan quota.
| Plan | Minimum interval |
|---|---|
| Free | 60 seconds |
| Standard | 30 seconds |
| Premium | 10 seconds |
| Enterprise | 10 seconds |
Changing the interval of an existing monitor takes effect on the next scheduled check — there is no immediate re-run.
Failure conditions
An HTTP monitor records a failure when any of these occur:
- Response HTTP status is outside the expected range (default: 2xx)
- No response within the configured timeout (default: 30 seconds)
- SSL certificate is expired or invalid
- Response body does not contain the required keyword (when keyword check is enabled)
- SSL certificate will expire within the configured warning threshold
Alert threshold
The alert threshold is the number of consecutive failures required before an incident is opened. This prevents transient network blips from waking up your on-call team.
- Threshold 1 — incident on first failure (recommended for critical paths)
- Threshold 2–3 — incident after N consecutive failures (use for less-critical services)
Once an incident opens, it stays open until the monitor recovers (all checks pass) or a team member manually resolves it.
Monitor status
| Status | Meaning |
|---|---|
| Up | All recent checks passed |
| Degraded | Some checks failed or response is slow (see below) |
| Down | All checks failing — incident is open |
| Paused | Monitor is manually paused; no checks run |
Degraded state
A monitor enters Degraded when it is technically reachable but performing poorly. This happens when:
- Response time exceeds the configured slow threshold (for example, above 2 000 ms)
- Only some prober regions succeed while others fail (partial regional outage)
- The SSL certificate is within the expiry warning window but has not yet expired
Degraded is an intermediate state between Up and Down. You can configure a separate, lower-priority notification channel for degraded events so your team is informed without being paged.
Prober regions
Checks are sent from geographically distributed prober nodes. Each check result is tagged with its originating region.
| Region | Location |
|---|---|
| EU West | Germany / Netherlands |
| EU East | Czech Republic / Poland |
| US East | Virginia |
| US West | Oregon |
On the Free and Standard plans, monitors run from a single region (EU West by default). On Premium+ plans, you can assign multiple regions to a single monitor. Checks then run in parallel:
- All regions fail → status: Down
- Some regions fail → status: Degraded
Correlation groups
Correlation groups let you link related monitors so that simultaneous failures produce a single correlated incident instead of multiple individual alerts.
Example: if your API, database, and frontend monitors all go down at the same time (likely a shared infrastructure failure), one incident is raised rather than three. Configure groups under Monitors → Correlation Groups.
SSL monitoring
For HTTPS monitors, SSL monitoring checks the validity and expiry of the TLS certificate on every probe cycle. You configure two thresholds:
- Warning — alert N days before expiry (default: 30 days)
- Critical — alert N days before expiry (default: 7 days)
SSL warnings appear in the monitor's incident history but do not count as outage time in uptime calculations.
Related
Was this page helpful?