Chaos engineering for serverless

on YouTube ↗34m

" summary

Shows how to form serverless hypotheses and actually run them: use AWS Fault Injection Simulator to change Lambda reserved concurrency to zero and auto rollback on CloudWatch alarms, or use SSM automation and a put-parameter action to flip Parameter Store flags. Demonstrates code-level fault injection with failure-lambda and chaos-lambda to add 100–400 ms latency, return 404 on ~50% of requests, or deny S3/Dynamo calls, and confirms experiments run and revert automatically.

AI-written summary · about

" chapters (12)

1. 0:00Chaos engineering intro
2. 1:04What is serverless
3. 3:54Serverless weaknesses
4. 6:22Fault injection techniques
5. 8:12Serverless app setup
6. 10:40AWS FIS config manipulation
7. 13:07Live config demo
8. 16:38Failure Lambda code
9. 19:32Live code demo
10. 25:08Automating experiments
11. 29:57Stop conditions
12. 32:51Summary

" related talks

• 43mApplying the Principles of Chaos to ServerlessYan Cui · SREcon18 Europe/Middle East/Africa
• 44mApplying principles of chaos engineering to serverlessYan Cui · Code Mesh LDN 2018
• 49mPerforming Chaos in a Serverless WorldGunnar Grosch · Agile India 2020
• 24mBreaking Serverless Things on Purpose: Chaos Engineering in Stateless EnvironmentsEmrah Samdan · Chaos Conf 2020
• 35mImprove resilience with automated chaos engineeringGunnar Grosch · Agile India 2021
• 29mImprove Resilience with automated chaos engineeringGunnar Grosch · Conf42 Site Reliability Engineering 2021
• 26mA proactive approach to resilience in modern appsGunnar Grosch · SREday San Francisco 2025 Q2
• 42mChaos Engineering: Why Breaking Things Should be PracticedM. Bonetto · Voxxed Days Trieste 2022