For mac OS, please install python3, virtualenv, docker, docker-compose. You can follow link below before start:

• python3 and virtualenv
• Docker
• Docker-Compose

Create file under attacker/task, for example atreus.py. Then create data in attacker/data, for example create atreus.py. After 2 files created, then edit file task you created previously attacker/task/atreus.py, then import your data from data.atreus import STATIC_DATA.

After attacker created, please create service exporter config here nginx/access-log-exporter/lite-atreus.yml. Edit the file base on your service/path requirements.

Execute command below to start all requirement services, for example if you want to start atreus:

make run-dev SERVICE=atreus

After execute command above you will have port open in 9090, 9100, 9093, 8080, 3000, 80, 4040. So make sure that, the port does not open before this service up. And oh, I think you want to know, why we need open too much port. It will be amazing to have all of those service, trust me!.

Remember that, all of the port can be open in localhost, for example Grafana. With user (admin) and password (foobar) You know all of the basic now, then, we can continue to the next phase, below!.

After all require services up, you need to boot the attacker with command below:

make prepare-attack
make attack-me SERVICE=atreus

make prepare-attack used to setup virtual environment for your python. It will isolate the python module for this project from globals.

This is the cool stuff, attacker is, the tools to help you make a load test to the service and make the grafana dashboard looks alive. After execute command above, you'll be have awesome locust dashboard under http://*:8089. Open the locust dashboard and input some data to make load test to the service.

After services and attacker up, it will create some important file to help you investigate when facing issue. Focus on the folder log, there are have some files which can help you to debug when facing issue/error.

• log/access-log-exporter/access-log-exporter.log (if rps does not appears in grafana)
• log/nginx/access.log (nginx access log)
• log/nginx/error.log (if nginx got issue)
• log/supervisor (supervisor/manager log)

Before start all of the service, you can create your own dashboad from provisioning folder grafana/provisioning/dashboards. Create your own json, or just copy paste from the existing and edit some related data, you know what I mean la....

Or if you want to create your ownn dashboard from scratch, just try use this simple query to collecting the rps:

sum(rate(lite_zeus_http_response_count_total[5m]))

If you facing issue when running this service, visit me at Terpusat

• Prometheus
• Locust