Automated Continuous Deployment at Heroku

I like posts like this. They share where they were, why they changed, how they did it, and were honest about the time it's taken to get there.

> Incident. This check fails if there’s a relevant incident open on the Heroku status site.

Curious, what's a relevant incident in this context? (given you are using the PaaS you are developing) Like what if you are deploying an emergency fix to a service that is in the middle of an incident?

> GitHub. The deployer will only promote changes which are present on the master branch.

I've always preferred the pattern Github uses internally of deploying feature branches and merging when deployment succeeds. Otherwise, there's weird behavior that requires manual intervention to fix in some edge cases, since deployments can fail, and it ends up cluttering the git log with operational noise.