If you want to avoid having to cycle the DB (i.e.: /etc/init.d/postgres restart), you can do this on the command line (as root):
kill -HUP `pidof postgres`
Once you do that, you will notice a message like this in the /var/log/postgresql/postgresql-9.4-main.log:
“2017-01-19 15:33:11 PST [24237-2] LOG: received SIGHUP, reloading configuration files”
Postgres receives the “hang up” (HUP) signal from the kill command and it knows to re-read the configuration files without taking the database down – ZERO outage.