Persistent SRE Antipatterns: Pitfalls on the Road to Creating a Successful
summary
Argues that alerts must target user-level service health not individual hosts, and that noisy host alerts cause alert fatigue; recommends automated remediation (example: scripted Cassandra ring repair) and velocity/threshold rules (ten node failures in three minutes should page). Promotes immutable infrastructure — bake images or containers, avoid config-management as a crutch — plus canary deploys and automated rollback tied to business metrics. Uses SLOs/error budgets, chaos engineering and customer-integrated metrics to trade off reliability and features. No physical Mission Control, on-call <50% ops time.
AI-written summary · about
chapters (41)
- 0:00Intro
- 1:41Service outages
- 2:55First alerts
- 4:02What makes a good alert
- 5:04Better tools
- 6:25How many alerts
- 8:04War rooms
- 9:20Sharing data
- 11:11Getting off the soapbox
- 12:54Oncall rotation
- 14:26Mission Control
- 15:25Reliability Theater
- 16:44Banners
- 17:03Something big goes wrong
- 17:25Its satisfying
- 17:39Incident management protocol
- 18:46Traditional infrastructure treats servers as pets
- 19:12Throwing servers off a roof
- 19:57Virtualized servers as cattle
- 21:05Containers vs Cattle
- 21:48Containers are moveable
- 23:01Puppet vs Chef
- 23:42Configuration Management
- 23:57Use the Company
- 24:30Configuration Management doesnt scale
- 25:01Automation
- 25:11Scripting
- 26:16Design Systems
- 26:56Integration
- 27:44Automate Repair
- 28:23Implement
- 28:46Rollback
- 30:16Force Multiplier
- 31:21Embedded SRU
- 32:29Production Ready Checklist
- 33:49Frameworks
- 34:53Nines
- 37:15SLOs
- 40:39Dust
- 42:28Customer Perception is Reality
- 43:40Integration with Key Customers
related talks
52mPersistent SRE Antipatterns: Pitfalls On the...
24mHow to Destroy an SRE
33mWhat are the common Anti Patterns of SRE?
25mUnleashing the potential of SRE for modern operations
18mDeploying Humans at the Edge of SRE
29mThe Curious SRE: Cultivating a Mindset for System Resilience
39mVirtuous Cycles: Enabling SRE via automated feedback loops
58mEnterprise Roadmap for SRE