If the machine has enough power (which for a Pi is heavily varying between generations), yes, you absolutely can! Look into something like Docker Compose (or even running NixOS).

Nah. Totally impossible. Give it up.

/sarcasm if not obvious

I have a Pi 4 that is running a bunch of stuff, including AdGuard(PiHole alternative), JellyFin, Radarr, Sonarr, Lidarr, Readarr, qBitorrent and some others.

I used DietPi as the OS. Would recommend as it has an AppStore like experience, with pre-configured versions of things ready to install.

DietPi is such a game changer.

Currently running it in a VM just for PiHole, and the ease of installing anything is nice.

Up until the point the resources available (CPU/ram/storage) become a problem.

That said, managing those apps as installations can be troublesome, which is why people like podman/docker. Updates are more contained, system updates don’t break services, and vice-versa.

A more comprehensive option is to use Proxmox or similar to manage services as a combination of Virtual Machines, LXC’S (Linux containers), and maybe podman/docker on a VM. Note that Proxmox techincally supports importing a docker container as an LXC, but this is a one-time action. Not recommended. (EDIT: For the record, there is a Proxmox on pi effort, but its community supported not Proxmox supported as of the last time I looked.)

Navidrome and pihole are the easy ones on that list also. Jellyfin itself is easy, but you aren’t going to be transcoding video on the fly, so make sure its in a format that is friendly to all of your devices, transcoding in advance if needed.

I used to have over 20 services running on a Pi4, I wouldn’t suggest running that much, but it’ll work

Depends on the Pi in terms of how much you can do, but yes.

I have a Pi 5, and I’m running Pi Hole, Nginx porxy manager, Jellyfin, Syncthing, Paperless NGX, and a full servarr stack (including Qbittorrent and a VPN). No issues; 4.6 gb out of 7.9gb ram in use, and no appreciable bottlenecks.

I would just caution that things like Jellyfin are a moveable feast - I am 1 user streaming content on my own to 1 device without issues at 4k. But more simultaneous users and if you start depending on transcoding your mileage may vary. So yes Jellyfin and other services will run fine, but any server can be put under pressure if it’s not adequate for the actual usage.

If you can run multiple applications, you can run multiple servers—a server is just an application with a network interface.

As other commenters have said yes you can run these depending on the Pi. Do note however if you need transcoding, A Pi is not recommended by jellyfin:

jellyfin.org/docs/general/…/hardware-selection#po…

Most Single Board Computers (SBC): Most SBCs (including the Raspberry Pi, especially the Raspberry Pi 5) are too slow to provide an acceptable Jellyfin experience as they often lack proper support for hardware acceleration. If you really want to run Jellyfin on an SBC, you may wish to consider models based on the following platforms: Rockchip RK3588 / RK3588S, Intel Core, Intel 12th gen N series

What is good enough fr transcoding? I have plex for travel, I want to move to jellyfin and get an arr stack all on one system. Looking at a 9th gen i5 thinkcentre with 16gb ram

Official Recommendations for low power: Intel 12th gen or above N series platforms Apple M Series Mac mini Rockchip RK3588 / RK3588S SBC (Advanced Users Only)

Note 10th gen and older integrated graphics are losing support in Linux due to Intel dropping SDK support or something. Don’t buy a nuc or small PC with Intel 10 or older but if you have one it’s fine for now.

Another option would be an old laptop with dedicated graphics (e.g. e waste/recyclers). then you can get away with older than 10gen Intel since you’d be using the external graphics. Just research how to limit battery charging in the bios since it’s always going to be plugged in.

I am looking at a few N100 mini pcs, is there an AMD alternative?

I keep seeing nearly good enough specs but the chip is not low power generation

it’s not recommended. AMD graphics in general have poor h.264 encoding: Encoder Quality: NVIDIA (RTX 50) ≥ Intel (Arc-B) ≥ Apple ≥ Intel ≥ NVIDIA ≥ AMD ≥ AMD H.264 (RX 9000) >>> AMD H.264*

Very good, I think I am going to go with a 16gb N100 lenovo all in one. Seems neatest.

Could I get your advice here on the below specs, I just came across it at a reasonable price for that much RAM. I would expect over a grand.

Intel Core Ultra 5 225H 64 GB DDR5 RAM 2 TB PCIe Gen 4 SSD

Based on those specs it sounds like it’s going to be more than capable in a homelab environment. If you’re streaming to ~5+ devices at the same time id assume it may start to struggle if it’s transcoding all of those so if you can pick or convert your media to something that doesn’t need transcoding that would be ideal.

but that looks like a solid buy. For reference I run on an old office potato and an external GPU from 2013 that I picked up for nothing a few years ago and have no problems. eventually I’ll have to upgrade when the drivers stop being supported but until then 2-3 streams max is fine for me.

I’ll be honest, I went looking based on your comment and found something thatnis either way under priced or very misleadingly titled. I have ordered it to check which.

It was 400 euro for that mini pc. I would probably move everything to it if it is real.

Define “server”:

If you mean any process that serves on the network well then of course you can.

If you mean multiple containerized stacks acting as a service then yes, you can do that too.

If you mean multiple operating systems, I’m not so sure but I can’t imagine that it’s impossible.

Depending on how you define multiple operating systems, you can use proxmox

Install a hypervisor and you can install as many operating systems as your hardware can support.

multiple operating systems, I’m not so sure but I can’t imagine that it’s impossible

The keyword is virtualization.

Easily! My Pi runs PiHole and a print server and I’ll could probably run almost my entire stack in it if I wanted. Jellyfin would be the hardest due to transcoding CPU usage, but it’s probably ok for one stream at a time.

The other issue for me would be storage. Since I have a 32TB NAS I run everything that requires large storage on it, but a USB RAID attached to a Pi would probably work just fine.

The other issue is RAM. I rarely hit 8GB, but do use more than 4GB fairly regularly.

Yes yes you csn.

The old may was we would run these direct like an app on the pi.

People then came up with containers and isolated things a bit made them easy to share and configure. - dockers

Have a look for people with blogs about all the software they are running on their pi plenty of info on how.

Pi isnt the best at jellyfin server as it isnt the best at transcoding files but for direct playing without having to convert the format its enough.

So yourself a favor and skip the pi and look for an old server on marketplace (loud and electricity is expensive, but lots of cores) or a NUC if u want a couple servers (or multiple docker containers)

I would skip Jellyfin on the pi. Other than that, it’s all good.

I always see people saying the Pi is poor for Jellyfin, but I’m running it on a 5 4GB (alongside a fair few other Docker images) and haven’t encountered any really issues. I’m only streaming 1080p and never see any stuttering or bad transcoding.

That said, I do intend to go with a SFF PC in the future as I’ve pretty quickly hit the point where it’d be better, so OP I would skip the Pi as well.

Let me be a bit more specific. If you are doing a couple of light services + jellyfin, maybe you can pull it off. But if you have some critical latency sensitive applications + jellyfin, it can cause issues at random times.

Also, if you are ONLY hosting Jellyfin with no *arr stack, you should be fine. I made that mistake and my Pi went down to a crawl.

Test your services together, and based on that decide on the subset of services.

I’m hosting a full Arr stack, Jellyfin, Navidrome, Calibre-Web, plus others. The only significant slowdowns I saw were when I ran a RAM-hungry Minecraft Java server as well.

Woah. I ran it on my 4GB Pi and it was running very slow. Which Pi are you using? I eventually moved to a mini PC because I was afraid of a microSD card fail + I also had to run Minecraft 😝

Yes but you should buy 3 and cluster them

Just get an oldish SFF or USFF office computer cheap somewhere. It’s generally much cheaper and allows for upgrades if ram/ssd prices ever come down. Also somewhat new Intel CPUs are quite good for transcoding.

I have like 30 different services running on the same machine.

That’s why you want to use some container solution, I use docker for simplicity.

Also raspberry pi is not great for self hosting. Any cheap x64 machine will be much more powerful for the same price, and have less compatibility issues.

Do not buy a RPI for self hosting. Get a mini PC second hand. It is much much better. And to answer you question, yes you can host multiple services on a single server. Read about virtualization and start learning the ways of proxmox, also say goodbye to your loved ones :)))

Don’t waste your time and money on a pi for this. He a cheap mini PC. Like the types that go in the back of the secretary monitor at your dentist office. That pike of junk I’d usually much better than a pi. It even has a box and room for a hard drive.

Acronyms, initialisms, abbreviations, contractions, and other phrases which expand to something larger, that I’ve seen in this thread:

Fewer Letters More Letters DNS Domain Name Service/System LXC Linux Containers NAS Network-Attached Storage NUC Next Unit of Computing brand of Intel small computers PCIe Peripheral Component Interconnect Express PiHole Network-wide ad-blocker (DNS sinkhole) RAID Redundant Array of Independent Disks for mass storage RPi Raspberry Pi brand of SBC SBC Single-Board Computer SSD Solid State Drive mass storage VPN Virtual Private Network

[Thread #114 for this comm, first seen 27th Sep 2026, 15:10] [FAQ] [Full list] [Contact] [Source code]

A lot of people saying ditch the Pi, but it’s fairly viable depending on what exactly you want from it, so long as you get one of the newer models with plenty of RAM. Simply streaming media is relatively manageable for a Pi. It will most likely run Navidrome okay, and Jellyfin will be okay if and only if media doesn’t need to be transcoded, and you MIGHT have issues with 4K, not sure. If you stream to something like a smart TV or Fire Stick, those can typically only handle certain video formats, so if your media isn’t that type, either the server transcodes it, which the probably Pi can’t do fast enough, or playback fails. Music typically doesn’t require transcoding, so that’s why I think Navidrome should be okay. Pi-Hole is manageable and is, after all, expressly designed to be run on a Pi. It can also be run on standard hardware, so while you can do it on a Pi, you don’t need one to use Pi-Hole.

Now, that said, two main concerns about a Pi. The first is that you’ll probably eventually want to host and do more, and you may run out of resources and wish you’d just gotten a cheap mini PC instead. The second is that you really should get an external HDD or something to store as much as you can on, preferably including the OS. An SD card just is not meant to handle the constant I/O that comes with being the main drive for a full computer. It’ll eventually fail, taking all the data on it with it. I think the best way to run off an external HDD has changed since I used a Pi, so some research recommended, but what I did before was flash the EXACT SAME image to the SD card and HDD, then delete the boot partition on the HDD, expand the data partition to fill the HDD, and delete the data partition off the SD card. If you use the exact same image for both, it at least used to be that the partitions keep the same UUID, so it boots from the SD card as normal, and when told to find the other partition by UUID, it finds it on the other drive and works just fine.

One big reason why not to buy a Rqspberry Pi for Self Hosting is the price. If you want to run more heavy services (like jellyfin can be), you need the Pis with 8GB of RAM. Though those where already expensive before the sharp rise is RAM pricing. When you include a fitting power supply and a small cooler with fan (so that you don’t run into thermal throttle fast) you are probably already at like 110€ at least. For that money you can also get a used mini PC (those that companies use for their employees), that already has more power and offers better upgradebility (RAM, storage). Also for some services you might run into the problem, there is no version build for ARM devices (though that happens very rarely nowadays).

Though with the RAM pricing problem every computing device gets more expensive. Just include such mini PCs into your comparison to make a good decision.