Start a Discussion or join our Discord for support and to help out.

WARNING

As of 2023-04-04, there is a showstopper bug, due to upstream library dependency, for running Machinaris on RaspPi. See the Known Issue below. Hopefully a fix will be available soon.

Install

Machinaris runs well inside Docker on Raspberry PiOS 64-bit systems, including DietPi, when used as a Harvester, connected to another system running as the Machinaris fullnode/controller.

Launching

Using the "New Worker" Launch Wizard, available on the Workers page of your Machinaris WebUI, it's easy to launch Machinaris in Harvester mode on your Pi device, to farm your plots.

Place the docker-compose.yml you generated from https://www.machinaris.app/ into a folder on your computer, could be in your Home diretory, I recommend naming the folder machinaris. Then open a shell in that folder:

docker compose config
docker compose pull
docker compose up -d
docker logs -f machinaris

Browse to the Machinaris WebUI on your primary Machinaris system, click the 'Workers' page and you should see the new RaspPi worker reporting in after 10 minutes or so.

Troubleshooting

Worker Memory

One Machinaris user noticed that some of his RaspPi systems reported memory on the Workers page, while others did not. He tracked it down to an issue in the older Pi4 bootloader. After updating to very latest version, Docker now allowed Machinaris to gather the memory usage of the container and report it on the Workers page.

Known Issues

"No such file or directory"

Unfortunately, one of the libraries used by Machinaris is not available on the aarch64 architecture. As such, you may encounter this bug, preventing the launch of the Machinaris API process required for the worker to report in.

/chia-blockchain/venv/bin/gunicorn: No such file or directory 

Hopefully a fix will be available soon. Details in this issue.