$ tail -f /var/log/incidents
Post-mortems, failure reproductions, and controlled experiments.
Building systems is only half of SRE. The other half is investigating, reasoning under uncertainty, containing failure, and learning from it. These write-ups document how I approach systemic failure.
A controlled experiment reproducing a cascading failure across microservices when the inventory service experiences memory exhaustion.
Read Post-Mortem