For single board systems / low memory systems, like the raspberry pi, I recommend only running a single celery worker. Celery is the task/worker component that handles all the searching and downloading behind the scenes.
If you skip this step, celery will use every available cpu (ie. the pi v4 has 4 cpus), which I've found is too much for the raspberry pi to handle.
Create a file .env
at the top level of the folder and enter:
NUM_CELERY_WORKERS=1
Then bring everything back up again:
docker-compose up -d