Changed our deployment system in the backend. Also took the opportunity to run some software updates and set up a second backup system; total downtime 1h 15m.
_Brief_ explainer: Because we didn't notice Philomena had a deployment script, we built our own that would run backups, then _pull_ code from the dev repo and reboot. Philomena's deployment script is based on the server _pushing_ code triggering commands, including the ones that handled such things as database migrations and in-place restarts. Now we should be using Philomena's deployment code... hopefully.