Oh Dear

Oh Dear monitors every premium site from the outside (uptime, certificates, DNS, broken links, Lighthouse) and polls Evolve's health endpoint for the inside view. Evolve's own checks and dashboard are described on Health Checks; this page is the per-site setup in Oh Dear.

Setup per site

1. Uptime and performance: checked every minute from several locations.
2. Certificate health and DNS: replaces any in-app SSL check.
3. Broken links, mixed content, Lighthouse: daily crawls. Lighthouse alerts double as SEO and performance regression warnings.
4. Application health: point it at https://<site>/oh-dear-health-check-results. Copy the secret Oh Dear shows into the panel under Settings > Health > Oh Dear, or set OH_DEAR_HEALTH_CHECK_SECRET in the .env. Oh Dear must send it as the oh-dear-health-check-secret request header; a query parameter does not work. The route only exists while the secret is set.
5. Status page and scheduled reports: a branded status page for the client and the monthly report to their stakeholders.
6. Alert routing: mail, Slack or SMS escalation is configured in Oh Dear, not in Evolve.

How the endpoint behaves

The endpoint serves the results of the last scheduled run. Oh Dear alerts when a check flips, when it recovers, and when results are older than ten minutes. That stale alert is the watchdog for a dead scheduler or a crashed application. HEALTH_OH_DEAR_FRESH_RESULTS=true would run every due check inside each poll, including a real test mail per poll; leave it off.

Keys

Variable	Default	Effect
OH_DEAR_HEALTH_CHECK_SECRET	unset	Enables and protects the endpoint. Editable in the panel, masked.
HEALTH_OH_DEAR_FRESH_RESULTS	false	Run checks inside the poll. Avoid.

Sites without Oh Dear can mail failures directly, see Health Checks.

Verifying

curl -H "oh-dear-health-check-secret: <secret>" https://<site>/oh-dear-health-check-results

returns JSON with one entry per check. In Oh Dear, the application health view lists the same checks within a minute of the first successful poll.