The performance is indeed pretty terrible. Most stuff runs fine on my NUCs except nextcloud. Maybe throwing more hardware at it solves it though.

Currently setting up my own monitoring stack:

• Fluentbit to gather metrics/logs
• Fluentd to aggregate logs
• Elasticsearch as database
• Grafana for visualization

Not sure if it is inside the scope of the project, but does it also have an API?

I think it's a reference to this: https://knowyourmeme.com/memes/i-will-not-eat-the-bugs

Could you elaborate on syncthing hub and spoke?

The idea is that you go through this process before the test