I’m currently exploring a very similar dilemma. I’m working on standing up a k3s cluster in my homelab to start moving some services over there. I’ve been using pangolin for a little while and it works great for simple docker vm setups and such.

What I’ve ended up doing (and this is just experimenting so take it with a grain of salt) is disabling traefik on my k3s cluster (personally don’t like traefik myself and prefer caddy) and just have pangolin manage subdomains and host routing. Then I threw MetalLB on the cluster (based on personal familiarity) and ended up recreating the “docker vm” IP address scheme. I have MetalLB manage IP addresses, and assign services to a specific LB that corresponds to the IP I want the service to have. Then just pop in that LB IP and the services port into pangolin and voila.

All that said, I’m still not entirely satisfied with the solution. I personally prefer subdomain routing and auth to be handled by separate systems (e.g. caddy + authelia) but pangolin acts as a one stop shop.

Ah fair! I also ran into that same problem, hence why I stopped lol seems we’re in similar situations. I ended up giving up on Spotify entirely in favor of YouTube music (yt premium), and I scrobble everything to last fm. Slskd helped the discovery aspect but it’s the Wild West as far as metadata tagging and organization goes. I also miss Spotifys playlists but Jellyfin instant mixes help, although I’m trying to go back to dedicated albums and artists more. Music world is tough with this stuff and streaming has gotten too convenient, so I feel your pain

What about lidarr import lists? Connects to your Spotify and you can sync your saved playlists

I’ve noticed there’s been a couple slower weeks lately, think you’re onto something there