Slamd container

Table of Contents

• Building
• Running The Server
• Running the clients
• Configuration

This repository hosts containerized slamd distributed load generator.

Building

To build the containers, first clone this repository, then update the submodules and run either ./podman-build or ./docker-build, depending on your preferred container tool.

💡	The service supervision scripts are built for podman. Adjustment will be needed to run them under docker

Running The Server

To start the server, simply run ./podman-run or ./docker-run. You only want one instance of the slamd server. Once it’s running, you can check podman ps or docker ps to find out where the web port is (what is mapped to 8080 in the container). Connect to that port in a web browser using http://localhost:PORT and you should see the slamd server running with one client connected from 127.0.0.1

Running the clients

Since the clients have to figure out the proper ports, you can start them with the sv/slamd-client/run script, and it will automatically figure out the ports and start a client connected to the slamd server that’s already running. To start clients from other machines, use the cli options to set slamd server’s host (SLAMD_ADDRESS) as well as the 3 port specifications. See ./podman-run -h or ./docker-run -h for the option specifications.

Configuration

Configuration of the client and server can be done before building, in the upstream/ directory which is the slamd distribution. Further work will be done to this process to make runtime configuration possible and streamlined, probably using configuration mounted with -v.