Show HN: Failpulse, simple HTTP and TCP uptime monitoring

It is.

There are a lot of these kind of sites out there, with the (probable) leader being pingdom.

Writing something basic is pretty easy for this, with a decent work-queue and a few worker nodes around the world checking from multiple locations.

The hard part is the notifications via SMS, billing, and dealing with failures that are en route from your testing-agent to the real site. (i.e. False positives where you think things are down when actually they're not.)