Unlike network.target the network-online.target guarantees that the network
devices are online.
If you bind to 0.0.0.0, [::], [::1], and/or 127.0.0.1 only that is enough to
proceed. But in case a particular IP is needed, like ${COREOS_PUBLIC_IPV4},
we require any IP assignments to have completed before Caddy's start. That
is achieved by depending on systemd-networkd-wait-online.service (which is
scheduled before network-online.target, then, automatically).
Add systemd service file for caddy
Add some README with basic setup instructions
Explain how to view the service configuration
Add a note about permissions
Add a comment about run user and group
service->service unit
A systemd service can consist of different units. A unit configuration
file has the `.service` file ending which is a bit confusing, so please
be considerate if I'm confusing `service` and `unit` in the README
Fix typos/reword
Add contact information
Made a faulty assumption that virualhosts could share acme proxy handlers; turns out they can't without fumbling up the middleware configuration (middleware chains overlap and cross over into other virtualhosts)!
Added a -grace flag to customize graceful shutdown period, fixed bugs related to closing file descriptors (and dup'ed fds), improved healthcheck signaling to parent, fixed a race condition with the graceful listener, etc. These improvements mainly provide better support for frequent reloading or unusual use cases of Start and Stop after a Restart (POSIX systems). This forum thread was valuable help in debugging: https://forum.golangbridge.org/t/bind-address-already-in-use-even-after-listener-closed/1510?u=matt
