SRE Weekly - SRE Weekly Issue #237

They fully expected their deep-discount sale to drive traffic, but they didn’t expect their system to handle the increase in the way that it did.

Michał Kosmulski — Allegro

Pre-stop hooks, liveness probes, and readiness probes were key to smoothly transitioning their services from a home-grown container system to Kubernetes.

Oliver Leaver-Smith — Sky Betting & Gaming

The experience of responding to an incident can evoke emotions that run the gamut.

Mads Hartmann

Google has released course materials the first of a series of classes on NALSD (“non-abstract large systems design”). This first one is about a distributed Pub-Sub system.

Auithor: Jenny Liao and Salim Virji — Google

Usually, doing a post-analysis on an incident you were in is an anti-pattern because you’re likely to introduce bias. But sometimes, it can lead you to learn more than you would have otherwise.

Lorin Hochstein