Now that SeatGeek is crushin it, it has become apparent that we need some sort of server/service monitoring beyond simple healthchecks with pingdom.
This should be split up into multiple parts, but the idea is thus:
Collect metrics from various servers, including, but not limited to:
- disk space
- load average