Running Kismet in a Docker Container

These steps will create a Kismet image from scratch. The initial image creation process builds Kismet from source. This is a CPU and RAM intensive process. Depending on your computer this can take a long time (hours). In my testing, a Raspberry Pi 4 will take several hours; a BeagleBone Black will take a 36-48 hours.

Swap File Requirements for Low RAM Systems

If your system has only a small amount of RAM (i.e. low RAM VPS', Raspberry Pi, Beaglebone, Gateworks, LattePanda, or other SBCs) you need to create a temporary swap file to be used during the initial image build. This is a one-time* requirement. Once the image is built, you do not need a swap file to run a container.

If you need a swapfile you can use make-swapfile.sh to create a temporary swapfile. You do not need a swapfile to run the container, just to build it.

Steps

• Clone this repo.
• Edit config/kismet_site.conf to your liking. Mostly this means specify the capture interface you prefer.
• Set permissions on logs so container can write to it (chmod 776 logs)
• Run docker compose up -d.

Your folder/file structure should look like this:

$ tree kismet
kismet
├── compose.yaml
├── config
│   └── kismet_site.conf
├── Dockerfile
└── logs

Kismet's log file are written to logs/.

---

Kismet with GPSD Stack

The folder kismet-with-gpsd/ contains Docker files to build a container stack that runs both Kisment and GPSD. You will need to edit the .env file with your GPS info. You also need to update kismet_site.conf with the preferred Kismet settings you want to use.