After a production outage, you're responsible for documenting what happened and making sure the team learns from it instead of repeating it.
How do you write post-mortems after outages?
