AMUD Dashboard Blog

I built AMUD. I'm biased. Here's the honest comparison anyway.

Deployed AMUD Dashboard behind Nginx, opened the page, everything looked fine except the CPU bar was stuck at 0% like the server was dead. Server wasn't dead. Nginx just ate the WebSocket upgrade.

Your dashboard is a map to your entire homelab. Treat it like one.

The default AMUD Dashboard look is orange glass. I like it. My partner said it looked like a Cheetos-themed spaceship cockpit. Fair.

I had a Home Assistant card that was literally just a URL. Useful, but my wall tablet deserved better.

"Is anyone watching something right now or can I restart the container?"

Not everyone runs Proxmox. Some of you are Docker-all-the-way-down and that's fine.

A link list tells you where Jellyfin should be. It doesn't tell you Jellyfin is stopped because you updated the LXC last night and forgot to bring it back up.

Here's a pattern I've seen in a lot of homelab monitoring glue:

I lost a Saturday to a YAML space.

Docker is fine. I use it. But wrapping a dashboard — a thing that mostly displays links and polls metrics — inside another network stack on a Proxmox box always felt like wearing two coats indoors.

My homelab had the usual problem: twenty services, zero good way to open them.

I tried the usual suspects. Heimdall looked nice until I checked htop and saw PHP-FPM doing absolutely nothing useful while holding ~150MB. Homepage was closer to what I wanted — live widgets, clean UI — but I spent an entire Sunday debugging a YAML indent before I admitted I hate editing config files on disk for something as dumb as "add Jellyfin link."

So I did what any reasonable person with too much free time does: I wrote another dashboard.