This isn't strictly necessary, this is just artifacts of the upgrade
process I did—and I figure may as well leave them here to help me in
the future if I need to do it again!
I'm not done with this yet, because I don't actually totally know how to upgrade the Postgres database files?
I could potentially just reset, since I don't care a *ton* about historical data, but I'll try to learn how `pg_upgrade` works in the context of Docker.
I tried to copy this change to the main Dress to Impress apps, and it
like SUPER broke them, even though I've seen sources say this should
work for both IPv4 and IPv6?
Clearly I'm missing something, but this was the change to get things
working again over there, so I'm gonna copy them over here too for good
measure.
Oh wow, TIL you need a special invocation in nginx to listen on IPv6 as
well as IPv4. This was both presumably breaking clients trying to
connect over IPv6 (I guess we never ran into that in a browser?), but
also breaking certbot's certificate renewal attempts, because Let's
Encrypt prefers IPv6 when possible. Okay!
I've learned some more about Ansible, and how to use `ansible.cfg` to
set up `ansible-playbook` to do the same thing as our `run.sh` was
doing! Same number of files, less overhead for the workflow's weirdness.
