Live: uptime of the public websites (Cron Trigger)

Goal. Up/down for setkernel's public websites on live.setkernel.com, checked from Cloudflare.

Work

A Cron Trigger fetches each public site's home page every minute, with a timeout: setkernel.com, raghavgupta.ca, hautedogfinders.ca, fourseasonsfarm.ca, programmes.net, and live.setkernel.com itself. It records up/down and latency.
Keep the last result and a 24 h up percentage in the Durable Object. Show them as small tiles; include them in /api/state.
Only public sites, from a list in wrangler.jsonc vars. Never internal hosts.
Tests: the result shape, timeout means down, and the 24 h percentage.

Done when npm run validate passes.

Decision: setkernel/hq#113 (Raghav's answers in its comments). Tracking: setkernel/hq#105.

Held until the redesign (#17) merges, so the uptime display follows the new design. Claude will put agent-ready back then.

Still on hold (agent-paused stays). The reasons have changed:

A Cron Trigger adds Worker CPU, and the account's Workers CPU is projected near its monthly allowance. Per-script attribution comes first.
The showcase direction isn't picked yet (Live v4: the showcase that wins customers (prototype first) #52).
hq#251's outside probes may cover the same uptime need.

It's released when the CPU cuts are in and #52 is picked.

— Claude Lead

Navigation Menu