Opbeans Load Generator

Summary

This is a load generator for Opbeans as a Docker container.

It has two primary modes of operation -- either as a stand-alone container which can generate load based on environment variables present at runtime or as a long-running web service which can receive commands to control load-generation over a HTTP REST interface.

Load is generated through invocation of molotov which is a load-generation tool provided by the Molotov project.

Standalone mode

Running

Start the docker container with a list of base URLs to generate load on

> docker run --rm -e OPBEANS_URLS=opbeans-node:http://opbeans-node:3000,opbeans-python:http://opbeans-python:3000 opbeans/opbeans-loadgen

HTTP mode

The use of HTTP mode was developed specifically as a load-generation component for use in the APM Integration Test suite.

Specifically, it is designed for use when the suite is launched using the --dyno flag, which enables user-controlled load-testing with the ability to manipulate the performance constraints of various services at-will.

Running in HTTP mode in a container [Recommended]

To run the Opbeans Load Generator in HTTP mode, start the container by defining the environment variable WS in the context of the running container. Typically, this is done using the -e flag as shown in the Docker documentation.

Note on running in the APM Integration Test Suite

When running the Opbeans Load Generator in the APM Integration Test suite with the --dyno flag, the WS environment variable will be defined automatically and the container will start in HTTP mode.

HTTP method overview

The list of HTTP methods and their purpose can be found in the http method overview section.

Further discussion

For further discussion, please see the detailed documentation in the README that is included in the /dyno directory.

Testing locally

The simplest way to run all tests for this project is by running:

> make test

Types of tests

BATS

Tests are written using BATS under the /tests dir.

Pytest

To run the Python tests, one must first install the test requirements:

> pip install requirements-dev.txt

This project includes a pytest test suite which provides a set of unit and functional tests.

To run these tests, both pytest and the pytest-flask modules must be installed and available to Python.

Publishing to Docker Hub locally

Publish the docker image with

> VERSION=1.2.3 make publish

NOTE: VERSION refers to the tag for the docker image which will be published in the registry