Self Healing Systems
summary
Explains how Rollbar and error monitoring create actionable signals that trigger automatic rollback or flip feature flags to restore service without human intervention, using ML fingerprinting to group exceptions and fire circuit‑breaker style responses. Requires cattle servers, blue‑green deploys and real‑time service checks, plus dynamic scaling and scheduled capacity increases. Claims SaaS control enables this level of automation and cites large teams deploying at high frequency — e.g., companies that deploy thousands of times per day.
AI-written summary · about
chapters (15)
- 0:00Self-healing systems intro
- 1:04Speaker and Rollbar mission
- 3:33Levels of system healing
- 6:04Self-healing requirements
- 7:28Self-healing characteristics
- 9:55Predicting system demands
- 10:57Error tracking core
- 12:23Crude vs effective detection
- 14:30Circuit breakers response
- 16:57Monitoring limitations
- 18:20Developer workflow evolution
- 20:44Error grouping ML
- 21:47Rollbar enables self-healing
- 23:53Handling customer behavior
- 24:55Summary
related talks
40mSelf-Healing Systems
34mToward continuous resilience
14mPlatform Engineering with Self-Healing AWS Infra
28mPanel: From Signals To Safety Nets: Self-Healing Progressive...
14mFrom Dashboard to Defense: Automating Resilience at
41mRestarts and rollbacks don't fix everything
40mWhen Incidents Fix Themselves: AI SRE in action
33mSelf-Healing Rollouts: Automating Production Fixes With Agentic AI