Toward a Culture Where Everyone Is Responsible for Outages
summary
A retrospective on transforming a product-focused development team into a cloud service team running thousands of instances, describing practical lessons and what had to be (un)learned. Covers cultural changes, system design around production code, impacts on architecture, creating Standard Operating Procedures, defining SLAs/SLOs, blast radius, noisy neighbors, SRE practices and risk management so teams take responsibility for service availability.
AI-written summary · about
chapters (13)
- 0:00Intro
- 3:19Context
- 8:12Difficulties
- 10:01From the system limits
- 13:15From the behavior expectations
- 15:38What is derived from the SLOS
- 20:26Start in the architecture
- 22:34Host your service early
- 26:12Sidebar: Site Reliability Engineering
- 29:06Observability
- 31:27Rollback / rollforward
- 33:47Standard Operating Procedures
- 37:14Pager transition ceremony
related talks
28mTowards a culture where everyone is responsible for downtime
25mBuilding a highly scalable system to assure Business Resilience
26mMission Critical Mindset Building a Culture of Resilience in DevOps Teams
54mCloud Native Culture
24mHow to survive Cloud – An Ops perspective
44mDeveloping a Culture of Reliability and Resilience through Chaos
29mThe Curious SRE: Cultivating a Mindset for System Resilience
16mCreating a Learning Culture