Fear and Loathing in Systems Administration

on YouTube ↗37m

" summary

Argues that operations pain is social not technical and prescribes concrete fixes: stop hoarding access, adopt pull requests for config, centralize logs with ELK or spunk and give teams the dashboard URL, add developers to oncall so the obvious 5:1 developer to systems ratio pressure eases, build small automation incrementally, replace shell log grabs and “terminal as a service” with a button via Windex, and measure outages and toil in money and time to make staffing requests credible.

AI-written summary · about

" chapters (4)

1. 0:00Start
2. 5:56Typical Day
3. 11:43Configuration Drift
4. 18:00Describe Everything That You Support

" related talks

• 34mDevops or: How I Learned to Stop Worrying and Love the Ops TeamMike Crute · PyOhio 2012
• 01h 39mPostOps: A Non-Surgical Tale of Software, Fragility, and ReliabilityTodd Underwood · LISA '13
• 05mOut of the Basement...Jamesha Fisher · devopsdays Chicago 2016
• 24mOperationalizing Observability: The Invisible PartsHazel Weakly · Observability Day 2025
• 33mDevops' Seven Deadly DiseasesJohn Willis · DevOpsDays Seattle 2019
• 33mOperations: the last Mile for DevOpsDamon Edwards · DevOpsCon Spring 2019
• 21mLightning Talks - Fear of the Bus: Docs and DevOpsHeidi Waterhouse · NDC Minnesota 2018
• 41mLessons from Mission-Critical Systems in Fast-Moving IndustriesNiek van Raaij · DevDays || DevOps Pro || CyberWiseCon Europe 2025