Uptime
Monitors check that your sites and services work, from several regions, and open an incident only when more than one place agrees something is wrong.
Monitor types
| Type | What it checks | Useful options |
|---|---|---|
| HTTP(S) | A URL answers with an expected status code. | Expected status codes, method (GET, HEAD, POST), request headers, maximum latency. |
| Keyword | A page loads and contains, or does not contain, some text. | Keyword, invert to alert when it appears. |
| API | An endpoint returns the right JSON. | POST body, JSON path and the value it must equal. |
| TCP port | A host accepts a connection on a port. | Host and port. |
| Ping | A host answers. | |
| DNS | A record resolves to what you expect. | Record type (A, AAAA, CNAME, MX, TXT, NS) and expected value. |
| SSL certificate | A certificate is valid and not close to expiry. | Minimum days remaining. |
| Domain expiry | A domain registration is not close to expiry. | Checked every 12 hours regardless of interval. Minimum days remaining. |
| Heartbeat | A scheduled job pings Bermward on time. | Expected interval and a grace period, 5 minutes by default. You get a unique ping URL. |
Targets that resolve to private, loopback or link-local addresses are refused, so monitors cannot be used to probe your internal network.
Intervals and regions
Check intervals available: 30 s, 1 min, 2 min, 5 min, 10 min, 15 min, 30 min, 60 min. Faster intervals need a higher plan. Probes run from us-east, eu-west, ap-southeast, and a monitor can be limited to some of them.
| Free | Starter | Growth | MSP | |
|---|---|---|---|---|
| Monitors | 3 | 25 | 50 | 250 |
| Fastest check interval | 300s | 60s | 30s | 30s |
How Bermward decides a monitor is down
- Every live region reports its latest result. Results older than three intervals (at least three minutes) stop voting.
- With two or more regions voting, two failing regions mean down. One failing region means degraded.
- If only one region is voting, it needs two failures in a row before the monitor goes down.
- A degraded monitor that was already down stays down until every region agrees it is up.
- When a monitor goes down, one incident opens. When it is up again, the incident resolves and the alert says how long it was down.
- Domain and heartbeat monitors are judged on a single server-side result, so one missed ping or one expiring registration is enough.
Maintenance windows
Schedule a window for chosen monitors, or for all of them, up to 14 days long and not already over. No incidents open for those monitors while it runs. Deleting a window re-evaluates the affected monitors straight away.
Suggested monitors
Bermward proposes monitors from what it finds during scans: your web hosts, certificates and domain registrations. Accept the ones you want in one step, or dismiss them. Accepted suggestions respect your monitor limit and the fastest interval your plan allows.
Getting alerted
- Chat channels and webhooks receive
uptime.incidentevents for incidents opening and resolving. See Webhooks and alert channels. - Admins can add up to ten phone contacts for SMS or voice alerts. These use alert credits, which are bought in packs on the billing page. Phone alerts need to be enabled on the server.
- On-call schedules decide who gets paged. See Organization and alerts.
API
| Method | Path | Role | What it does |
|---|---|---|---|
GET | /api/orgs/:orgId/uptime/monitors | viewer | Paginated list. Query: q, page, pageSize, state. |
POST | /api/orgs/:orgId/uptime/monitors | analyst | Create a monitor. |
GET | /api/orgs/:orgId/uptime/monitors/:monitorId | viewer | One monitor. |
GET | /api/orgs/:orgId/uptime/monitors/:monitorId/series | viewer | Response time and availability series. |
GET | /api/orgs/:orgId/uptime/monitors/:monitorId/checks | viewer | Recent raw checks. |
PATCH | /api/orgs/:orgId/uptime/monitors/:monitorId | analyst | Edit or pause. |
DELETE | /api/orgs/:orgId/uptime/monitors/:monitorId | analyst | Delete. |
GET | /api/orgs/:orgId/uptime/incidents | viewer | Incidents. |
POST | /api/orgs/:orgId/uptime/incidents/:incidentId/acknowledge | analyst | Acknowledge. |
POST | /api/orgs/:orgId/uptime/incidents/:incidentId/resolve | analyst | Resolve. |
GET | /api/orgs/:orgId/uptime/maintenance | viewer | Upcoming and past windows. |
POST | /api/orgs/:orgId/uptime/maintenance | analyst | Body: title, startsAt, endsAt, optional monitorIds. |
DELETE | /api/orgs/:orgId/uptime/maintenance/:windowId | analyst | Delete a window. |
GET | /api/orgs/:orgId/uptime/suggestions | viewer | Suggested monitors. |
POST | /api/orgs/:orgId/uptime/suggestions/accept | analyst | Body: keys. |
POST | /api/orgs/:orgId/uptime/suggestions/dismiss | analyst | Body: keys. |
Monitors can also be managed with an API token. See API.