Hacker News new | ask | show | jobs
by geocrasher 664 days ago
Honestly? Look at Netdata for comparison. Everything from nginx hostname requests (we run web hosting servers) to cpu/ram/disk data but also network data and more. If you can do better than that somehow, by all means do it and make it better.

But there's more to it than just collecting data in a dashboard. Having a reliable agent and being able to monitor the agent itself (fore example, not just saying "server down!" if the agent is offline, but detecting the server remotely for verification) would be nice.