« Settings

Idempotency »

Chaos

In the spirit of Netflix’s Chaos Monkey, this framework has the ability to inject failures into service calls to help simulate problems, and aid workflow authors in writing robust, idempotent services.

In settings(local).py add a dictionary called AWS_CHAOS. This dictionary maps AWS services to a dictionary of exceptions and approximate failure percentages (0.0-1.0). 0.5 means approximate 50% of calls to the given service will fail.

For example

PRIMARY_STREAM_SOURCE = 'arn:aws:kinesis:eu-west-1:999999999999:stream/aws-lambda-fsm'
PRIMARY_CACHE_SOURCE = 'arn:aws:dynamodb:eu-west-1:999999999999:table/aws-lambda-fsm.cache'

from botocore.exceptions import ClientError

AWS_CHAOS = {
    PRIMARY_STREAM_SOURCE': {
        ClientError({'Error': {'Code': 404, 'Message': 'AWS Chaos'}}, 'service'): 0.1,
    },
    PRIMARY_CACHE_SOURCE: {
        ClientError({'Error': {'Code': 404, 'Message': 'AWS Chaos'}}, 'service'): 0.1,
    }
}

This works locally and when deployed to AWS.

« Settings

Idempotency »