Load Generator

A tool for generating HTTP traffic against specified targets in infinite loop.

Configuration

All configuration options, with default values, are defined in Python file located within the project directory under the path:

./config/config.py

Meaning of individual options is following:

• REQUESTS - list of dictionaries - List of requests to send, consecutively in a loop. Every dictionary contains information about single request. It has to contain following items:
  • url - string - url to send the request to.
  • http_method - string - name of the http method to use when sending the request.
  • headers - dict - contains http headers to use with the request.
  • data - dict - URL params in case of GET request or JSON body in case of PUT or POST. Not aplicable in case of DELETE, HEAD and OPTIONS
• FREQUENCY - integer | decimal - number of cycles per second when the requests are being sent. To send requests less frequently than one second use a decimal.
• RANDOMIZE_FREQUENCY - boolean - if True, the delay between requests will be randomly decreased by the value between 0 and up to full delay.
• CONCURRENT - list of integers - number of requests sent concurrently, consecutively, during the next cycle.
• LOG_LEVEL - string - python logging framework logging level.

Use

The tool is designed to be deployed in k8s cluster and run until killed.

However, it can be also run from commandline by issuing a following command:

make run

Install project's dependencies (and dev-dependencies) using pip:

$ make init

To run the tests:

$ make test

To run the linter:

$ make check