Skip to content

Production Monitoring

Last reviewed: 2026-07-14

Project NEXUS uses external uptime checks plus application error tracking. Monitoring configuration should avoid committing private contact details, tokens, or webhook URLs.

External Checks

Configure HTTPS monitors for:

URL Type Expected result Suggested interval
https://api.project-nexus.ie/api/v2/health HTTP keyword Health JSON reports OK 5 min
https://api.project-nexus.ie/health.php HTTP status 200 when dependencies are healthy 5 min
https://app.project-nexus.ie/ HTTP keyword Project NEXUS shell loads 5 min
https://accessible.project-nexus.ie/ HTTP keyword Accessible frontend loads 5 min
https://project-nexus.ie/ HTTP keyword Sales site loads 5 min
https://app.project-nexus.ie/manifest.json HTTP status 200 15 min

Use at least two alert destinations for primary checks, such as an owner-controlled email address plus an incident channel. Keep private contact values in the monitoring provider, not in this repository.

Cloudflare Health Checks

Cloudflare health checks are useful as a second vantage point:

  1. Check api.project-nexus.ie at /api/v2/health, expecting HTTP 200.
  2. Check app.project-nexus.ie at /, expecting HTTP 200 plus a stable body keyword.
  3. Notify an owner-controlled alert destination on status changes.

When An Alert Fires

  1. Confirm the alert from a second source.
  2. Check the API health endpoint and the pre-framework health.php endpoint.
  3. Check the active blue/green color:
sudo bash scripts/deploy/bluegreen-deploy.sh status
  1. If the incident started immediately after a deploy, prefer rollback before deep debugging:
sudo bash scripts/deploy/bluegreen-deploy.sh rollback --detach
  1. If the platform is unusable and writes must pause, use scripts/maintenance.sh to toggle both maintenance layers.