Nothing is being checked. The scheduler last ran 4 days ago. It should run every minute. Add this one line to cron: Ask your host for the versioned CLI PHP path, then run php artisan pharos:cron with that PHP. The web PHP selector does not configure cron.

Bring in

Let your monitoring and jobs keep this page up to date on their own

Connect a source

Pharos changes the component; you decide what sends it

What knows whether it is healthy?
Pharos can also check websites and ports itself: set that on the component.
Which component does it update?
Only components without a built in check are listed, so two sources never fight.
Use a token with write access
The token proves the request comes from you. It only works for this page.
Write token Create a new token

A token is shown once, when it is created. Paste it where the examples say YOUR_TOKEN.
Copy this into your tool
In n8n add an HTTP Request node with these settings.
Request
PUThttps://demo-pharos.solutionmax.net/api/v1/pages/harbor/components/15
Header
Authorization: Bearer YOUR_TOKEN
Status to send

Body
{"status": 1}
The same as a shell command
curl -X PUT 'https://demo-pharos.solutionmax.net/api/v1/pages/harbor/components/15' -H 'Authorization: Bearer YOUR_TOKEN' -H 'Content-Type: application/json' -d '{"status": 1}'
Existing clients may also send the token in an X-Cachet-Token header.
This changes the component only. It does not open an incident or notify anyone.
Notification type: Webhook, method POST, body application/json. URL:
POSThttps://demo-pharos.solutionmax.net/api/v1/pages/harbor/integrations/kuma/15
Additional headers (JSON)
{"Authorization": "Bearer YOUR_TOKEN"}

Kuma sends every change on its own. Up sets Operational, down sets Major outage, maintenance sets Under maintenance, pending changes nothing. Kuma's test button sends no state and is refused; test with a real monitor.
Warehouse scanners: call this at the end of the job
```
curl -fsS -X POST https://demo-pharos.solutionmax.net/api/v1/pages/harbor/heartbeat/hb_harbor_scanners
```
Expected every 5 minutes. Change

No check in on time: the component goes to Major outage. The next check in sets it back. The secret in the URL is the authorization; no token is needed.
Need an incident with a message instead of a status change?
POST this to https://demo-pharos.solutionmax.net/api/v1/pages/harbor/incidents with the same header. It appears on the page and is mailed to subscribers. Keep the returned data.id and post follow ups to https://demo-pharos.solutionmax.net/api/v1/pages/harbor/incidents/INCIDENT_ID/updates; send status resolved to close it.
{ "name": "Service unavailable", "status": "investigating", "message": "We are investigating a service interruption.", "impact": "major", "components": { "15": "major_outage" } }

Already feeding this page

4 components

Pharos checksTracking portal (HTTP), Driver app API (HTTP) 2 running
Warehouse scannersHeartbeat · last check in 22 Oct 21:05 Live
EDI gatewayAPI · last change 22 Oct 21:06 Set from outside

The other direction

To tell n8n or a chat when an incident opens, add a destination under Send out.

Bring in

What knows whether it is healthy?

Which component does it update?

Use a token with write access

Copy this into your tool