SRE Weekly - SRE Weekly Issue #316

Yes.

  Chris Evans — incident.io

If you don’t test them, you don’t have backups; you have a lottery ticket. Except the chance of winning is high. And the prize is data loss.

  Emily Arnott — Blameless

Being blameless does not mean blaming no one outwardly and blaming yourself inside your head.

  Emily Arnott — Blameless

LinkedIn’s Alert Correlation system posts recommendations to Slack about which microservice may be at the heart of an incident.

  Nishant Singh — LinkedIn

I always get the two confused. This article explains the difference and gives tips for writing runbooks. More on runbooks from the same folks here.

  Jessica Abelson — Transposit

There are many intricate details in there! For example, the S3 SLA is per calendar month, not a rolling window, so the SLA of your product based on it might need to match.

  Alex Ewerlöf

The ins and outs of conducting an effective postmortem. Ready templates and examples from leading organizations around the world!

  Prathamesh Sonpatki — Last9