The biggest traffic spike I saw on http://www.tripfate.com was between 1-7 PM Eastern on 12/25... I really couldn't justify the traffic (had no ads or anything) so I started to investigate, and it looks like the majority of the traffic was coming from the other post I published on IH last week: https://www.indiehackers.com/forum/how-i-met-my-wife-and-why-i-built-yet-another-flight-price-tracker-34fb46049e
I'm really glad I used the nginx + php7-fpm stack in a Docker container to deploy on DigitalOcean. It took me literally 3 minutes to spin up another instance behind the load-balancer and I had my environment ready in seconds thanks to everything being Dockerized.
Reverting back to a single instance was equally easy - I'd highly recommend this approach rather than doing everything manually. Oh and you should totally leverage DO's monitoring service - it's awesome! I receive a Slack alert any time the resource consumption approaches 70%.
Although I still don't know what caused the traffic spike (except for maybe a bunch of people decided to read that IH post on Christmas), I'm really glad I had prepared for it.
The next thing(s) I'd probably like to build is automatic deployment through Jenkins with BitBucket, and triggering a script behind the scenes to deploy more instances behind the load-balancer once the cluster consumption reaches ~70%... but that's for another day.
Oh and I'm still looking for beta-testers for http://www.tripfate.com so please feel free to sign up! :)