KodeKloud Notes

In the world of Chaos Engineering, tools like Netflix’s Chaos Monkey pioneered fault injection testing. Today, platforms such as Gremlin, Azure Chaos Studio, and AWS Fault Injection Simulator (FIS) help teams validate system resilience under real-world failure scenarios.

AWS FIS is a fully managed service that lets you run fault-injection experiments on AWS workloads. By deliberately introducing failures, you can:

Identify weaknesses before they affect customers
Validate auto-scaling, failover, and recovery processes
Ensure SLAs are met under adverse conditions

Note

AWS FIS supports both simple and complex scenarios—from terminating individual EC2 instances to simulating an Availability Zone outage.

Key Benefits

Benefit	Description
Fully Managed	No need to provision infrastructure for fault injection
Native AWS Integration	Works with IAM, CloudWatch Alarms, AWS X-Ray, EventBridge, and more
Prebuilt & Customizable	Use built-in templates or define your own experiments
Multi-Environment Support	Inject faults in EC2, ECS, EKS, RDS, Lambda, and more

AWS FIS Architecture

AWS FIS integrates seamlessly with your AWS environment:

CloudWatch Alarms: Trigger experiments or remediation workflows when thresholds are crossed.
AWS X-Ray: Correlate faults with distributed traces to pinpoint failures.
EventBridge: Automate experiment scheduling and notifications.

You can leverage AWS FIS to simulate:

EC2 instance terminations and CPU/network stress
ECS and EKS pod failures
RDS instance failovers
Availability Zone outages
Network latency and packet loss between resources

Managing Experiments

AWS FIS experiments are defined as JSON documents. You can manage them through:

Interface	Command / Action
AWS Management Console	Create, configure, and run experiments via the web UI
AWS CLI	`aws fis create-experiment-template`
AWS CloudFormation	Use the `AWS::FIS::ExperimentTemplate` resource

# Example CloudFormation snippet
Resources:
  MyFISExperiment:
    Type: AWS::FIS::ExperimentTemplate
    Properties:
      Description: "Terminate EC2 instance for resilience testing"
      Actions:
        - ActionId: aws:ec2:terminate-instances
          Parameters:
            instanceIds: ["i-0123456789abcdef0"]
      Targets:
        - ResourceType: "aws:ec2:instance"
          ResourceTags:
            chaos-test: "true"
      RoleArn: arn:aws:iam::123456789012:role/FISRole
      StopConditions:
        - Source: "aws:cloudwatch:alarm"
          Value: arn:aws:cloudwatch:us-west-2:123456789012:alarm:HighCPUUtilization

Warning

Always run experiments in a staging or non-production environment first. Fault injection can cause service interruptions!

Security & Permissions

Leverage AWS Identity and Access Management (IAM) to grant granular permissions:

fis:CreateExperimentTemplate
fis:StartExperiment
fis:StopExperiment
cloudwatch:DescribeAlarms
ec2:TerminateInstances

Use IAM policies and roles to restrict who can create, modify, or execute FIS experiments.

Links and References

Watch Video

Watch video content