URLs Uptime Watch

Iterates over a list of URLs and fires a per-URL alert when any returns a non-success HTTP status code (≥ 400). One alert per failing URL per fire.

What it does

On each fire, the monitor's for_each: block walks the URL list. Per URL:

1. The check fires (http tool with method GET against {{url}}).
2. The when: arm tests {{check.responseCode}} >= 400. Any 4xx or 5xx triggers the alert.
3. Healthy URLs (response < 400) fall through to the else: arm and notify nothing.

So a 3-URL list with two healthy and one returning 503 produces ONE notification, not three.

Edit before installing

In monitor.yaml:

• for_each.items — replace with your URLs.
• for_each.as — the loop variable. Default url; if you rename it, update every {{url}} reference downstream.
• The condition >= 400 treats redirects (3xx) as success. If you want to also alert on 3xx for a specific URL, tighten with {{check.responseCode}} >= 300 && {{check.responseCode}} != 200.
• Schedule every 5 minutes for production canaries; every minute for high-touch endpoints (paying close attention to rate limits and noise).

Bindings used per iteration

• {{url}} — current URL from the list.
• {{check.responseCode}} — HTTP status code (number).
• {{check.responseBody}} — response body (string).
• {{monitor.firedAt}} — ISO 8601 fire timestamp.

Tools used

• http — core InTouch HTTP task type. Public endpoints need no credential; authenticated endpoints can set check.credentialName:.

Pattern showcase

This is the literal for_each over a list pattern, mirroring stocks-portfolio-watch but for URL health.

Compared to the single-URL api-health-check monitor: api-health-check watches one URL with a static check; this monitor handles many URLs with one schedule fire. Pick api-health-check for a critical endpoint that deserves its own monitor; pick this one for a fleet you want to bulk-watch.

Previous form

Replaces the website-uptime-check job (deleted 2026-05-19). The job version made three separate http tasks and relied on notifications.alertOnError: — this monitor uses for_each: for a cleaner shape and a per-failing-URL notification body.