This post contains resources for the session Chaos engineering for serverless with AWS Fault Injection Simulator.

Companies of all sizes and industries perform chaos experiments on instance- and container-based workloads. However, serverless functions and managed services present different failure modes and levels of abstraction.

This session looks at forming hypotheses to fit serverless, what the experiments can achieve, and how to perform them safely using AWS Fault Injection Simulator and open-source libraries.

Fault injection for AWS Lambda - failure-lambda

failure-lambda is a small Node module for injecting failure into AWS Lambda. It offers a simple failure injection wrapper for your Lambda handler where you then can choose to inject failure by setting the failureMode to latency, exception, denylist, diskspace or statuscode.
https://github.com/gunnargrosch/failure-lambda

Chaos Injection for AWS Lambda - chaos_lambda

chaos_lambda is a small library injecting chaos into AWS Lambda. It offers simple python decorators to do delay, exception and statusCode injection for your AWS Lambda function. This allows to conduct small chaos engineering experiments for your serverless application in the AWS Cloud.
https://github.com/adhorn/aws-lambda-chaos-injection

SSM Document for storing configuration in Parameter Store

Enables support for putting configuration in parameter store to enable AWS Lambda faults via libraries.
https://github.com/adhorn/aws-fis-templates-cdk/blob/main/lib/fis-upload-ssm-docs/documents/ssma-put-config-parameterstore.yml

AWS FIS experiment using SSM automation and AWS Lambda libraries

https://github.com/adhorn/aws-fis-templates-cdk/blob/main/lib/fis-experiments/lambda-faults/experiments-stack.ts

AWS Resilience and Chaos Engineering Day

Watch the recorded session from AWS Resilience and Chaos Engineering Day to learn more about chaos engineering and resilience.
https://pages.awscloud.com/EMEA-field-OE-resilience-chaos-engineering-day-202111-reg-event.html

Resilience and Chaos Engineering Community on Twitter

Learn from and connect with others who are using Chaos Engineering to improve performance, observability, and resilience.
https://twitter.com/i/communities/1496442468375703556

If you have any questions, please reach out on Twitter or LinkedIn. Go experiment!