Add health check service

[donkirkby]

We had a server mysteriously power off, and it wasn't noticed for a few days. Add a health check service that pings all our ongoing services around the network on a regular schedule. (Somewhere between 10 minutes and an hour?)

I think the best method is the --host option of systemctl, but healthchecks.io might also be useful. I guess it makes sense to configure a YAML file with a list of hosts and service names, along with some kind of schedule and notification lists. Some services might be more convenient to monitor through the modification date of a log file. I guess we'd need two copies on separate hosts to monitor each other.

Think through the security implications, because you need SSH access. Maybe configure a very limited account for health check access.