Documentation

Everything needed to get a monitor, an alert rule and a status page running. Last updated .

Quickstart

Create an API token under Settings → API tokens, then create your first monitor. Tokens are scoped per organisation and can be limited to read-only.

export PLAMB_TOKEN="plamb_pat_<token>"

curl -X POST https://api.91831082026.site/v1/monitors \
  -H "Authorization: Bearer $PLAMB_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
        "name": "storefront",
        "type": "http",
        "url": "https://example.com/healthz",
        "interval": 60,
        "regions": ["fra", "ams", "waw"],
        "expect": { "status": [200], "body_contains": "ok" }
      }'

A monitor starts probing immediately. The first result usually lands within one interval.

Check types

  • http — status code, response body, redirects, TLS chain and total time.
  • tcp — connect success and latency for a host and port.
  • icmp — packet loss and round-trip time.
  • heartbeat — an inbound ping your cron job sends; a missed window opens an incident.

A check is only marked down after the failure is reproduced from a second region, which removes most false positives caused by a single transit provider.

Status pages

A status page groups monitors into components. Components can be public while the underlying monitors stay private, so internal dependencies are not exposed.

To use your own domain, add a CNAME to pages.91831082026.site and add the domain in the dashboard. Certificates are issued and renewed automatically.

status.example.com.  300  IN  CNAME  pages.91831082026.site.

Alerting

Notification rules match on service, severity and time of day. Each rule points at one or more channels: email, Slack, Telegram, or a signed webhook.

Webhook payloads are signed with HMAC-SHA256 over the raw body. The signature is sent in X-Plamb-Signature and the timestamp in X-Plamb-Timestamp; reject anything older than five minutes.

HTTP API

MethodPathDescription
GET/v1/monitors List monitors with their current state.
POST/v1/monitors Create a monitor.
GET/v1/monitors/:id/results Raw check results for the last 30 days.
POST/v1/maintenance Open a maintenance window.
GET/v1/incidents Incident history, newest first.

Probe addresses

If you filter inbound traffic, allow the probe ranges below. They are also published as probes.91831082026.site in DNS and change no more than twice a year, always with 30 days notice.

RegionCodeUser agent
Frankfurtfra Plamb-Probe/1.4 (+https://91831082026.site/docs#probes)
Amsterdamams Plamb-Probe/1.4
Warsawwaw Plamb-Probe/1.4
Helsinkihel Plamb-Probe/1.4

Rate limits

The API allows 120 requests per minute per token. Responses carry X-RateLimit-Remaining and Retry-After. Bulk imports should stay under 20 writes per second.

Endpoint health for this edge node is exposed at /healthz and returns 200 ok while the node is serving traffic.