Looks like we have a few subscribers to this community so I figured Id make a “poll” of sorts. Are most people already rocking the servarr stack or do we have anyone who is here to learn? If you already have a stack, whats your setup? Ill go first.

Im currently running my homelab on consumer hardware, just an old gaming PC I had. i7-4790k with 32gb of ddr3 ram and a gtx 1060. Im running truenas Scale on it, and I moved from ubuntu server. Ive got the whole *arr stack running except for whisparr along with a bunch of other self host programs like vault warden and audiobookshelf.

  • I have everything running on an old r510 I got for free. It has 2xL5640 CPUs and 80gb of RAM. I am running unRAID with about 50tb of total storage, and Readarr, Sonarr, Prowlarr and Overseerr for handling requests. I also have audiobookshelf and calibre-web running.

    I started with windows server and moved to running plex on Ubuntu server when windows started having issues. Then I decided I wanted to virtualize it so I moved to Proxmox with a plex VM and a servarr VM both had access to zfs pool of 6x3tb HDD. Eventually I wanted to upgrade without losing all the data and without the budget to upgrade all 6 HDD at the same time so I moved to unRAID. Now everything runs in the built in docker containers.

    I love the convince of the *arr apps. I have automated almost every part of my media library including meta stuff like showing if a show is cancelled or ended with PMM. I am not a huge fan of readarr or lidarr because the last time I tried using them they had a hard time getting the metadata for the books and music. I ended up ditching my music collection, and I use calibre for handling my book collection now.

    • Readarr/Lidarr certainly aren’t as robust as the other *arr seem to be. I find myself always using ABS’ match function for books Im adding, but also part of that might be that Readarr only has the mouse website for a source right now and those files don’t always have good metadata/file structures. For my music I actually started using lidarr-extended as it allows deemix for a source. Works wonders for me.

  • Pretty new to self-hosting type stuff, so I went with a Synology NAS when building my setup this spring. Got a DS923+ and two Seagate 12 TB IronWolf drives, and for overkill, added an extra 4 GB of RAM (for a total of 8). There was a lot of fuss on the Reddit threads I browsed about the later Synologys being bad for Plex, but that wasn’t really a concern for me since I have Plex on an Nvidia Shield.

    I’ve got Sonarr / Radarr, with Bazarr for subs, Prowlarr for torrent index management, and Overseerr for day-to-day interaction with the stack. Using Portainer to manage all my docker containers, and Watchtower for auto-updating most of them. Downloads are through qBittorrent with gluetun for VPN. I was using Mullvad with port forwarding but now that’s going away, I’m probably going to switch over to Proton VPN. Not really looking forward to doing that reconfiguration.

    I wasn’t quite sure this was all worth the effort until everything lined up and I started submitting downloads through Overseer - it’s SO nice to send movies or tv shows to be downloaded through a snazzy UI.

    Future plans include setting up a self-hosted photo library browser, and some kind of self-hosted music setup, though most of my downloads are from Soulseek and nothing *Arr really supports that as a source last I looked into it.

  • I have a Dell R730XD with 128Gb RAM and a 49TB array. It’s way overkill for my needs, but I have fun with a lot of docker containers running on Unraid. I can also host a few web apps for my own needs. The Unraid docker GUI became way too slow to use, so now I manage everything through Portainer and modified everything to use stacks and docker compose. It’s much easier to manage now.