Selfhosted

Hello!

So until now, I’ve been self-hosting at home with my own machines, and this is the first time I’m working with a VPS.

I’m running Nginx Proxy Manager (NPM) and my application in separate Docker containers on my AlphaVPS VPS with the IP address 100.100.10.10. I’ve configured UFW to secure my server, and I’m trying to access my application through a domain using NPM, but I’m getting a 504 Gateway Timeout error.

Here’s what I’ve done so far with UFW:

1. Block All Incoming Traffic by Default

ufw default deny incoming
ufw default allow outgoing

2. Allowed Specific IPs

I’ve allowed specific IPs which are my home and work IP:

ufw allow from 100.100.10.11
ufw allow from 100.100.10.12

3. Allowed Ports for Nginx Proxy Manager

I’ve opened the necessary ports for HTTP (80) and HTTPS (443) to be accessible from the outside:

ufw allow 80
ufw allow 443

What I did in Nginx Proxy Manager:

• I created an A record for sub.domain.com through my registrar, pointing to the VPS IP (100.100.10.10).
• In Nginx Proxy Manager, I added a Proxy Host for the domain sub.domain.com, set the IP address (100.100.10.10), and the port of my application inside Docker (e.g., 8000).
• I also enabled SSL and requested a new certificate for sub.domain.com.

Issue:

Even after setting up everything, I’m facing a 504 Gateway Timeout error when trying to access my application through the domain name.

Could someone help me troubleshoot why I’m getting the 504 Gateway Timeout error? I might have done something wrong as my understanding of this is very basic. Could it be a problem with the UFW setup, or is there something wrong with the Nginx Proxy Manager configuration?

Thanks in advance for your help!

Hey, folks!

So, the card linked to my hetzner account expired, and while I updated it everywhere else, the hetzner info fell through the cracks. They deleted my storageshare server, and erased everything, because I'm a fool, and didn't have their emails going to an account I ever check. I've spent the last several days feeling like I'd had a digital housefire. Things kept popping into my head, photos I had taken 10 years ago, or early drafts of the novel I'm writing... It's been pretty fucking depressing.

But, tonight, I fired up a laptop I haven't used in a while to find that most of what was in nextcloud was backed up on it. It's not everything, but it's the bulk of it.

I'd like for this to never happen again. I'm wondering if there's a complete idiot's guide to self hosting nextcloud? When I say I know nothing about this, please believe me. We're talking starting from scratch. I've never self hosted anything, and I have no idea where to begin. I'm on fedora silverblue, but just because I'm using linux doesn't mean I know anything. It just means I'm cheap. Haha. All I know is that I never want to go through that feeling of complete loss again. I'll make sure that whatever I do, it'll be backed up in two locations at least. I was paying for the family plan, and my brother, his wife, my mom, and a friend lost access to their stuff, too. So far as I know, there isn't a back up of their stuff. I really messed up here.

Any help is really appreciated, thanks in advance!

So here's the state of things right now:

I've got a Synology NAS with stuff like my movie collection stored on it. I think at some point I'll move my music collection up there, too. It's an ARM powered 2-bay thing, I'm not particularly interested in using any of Synology's software, and there's some stuff that won't run on that box because it's ARM instead of x86. I don't really have a "server" box running.

A few years ago I bought a "commercial TV" aka one that doesn't have Roku or whatever. For awhile, I ran OSMC (basically Debian Kodi) on a Raspberry Pi, which...I had enough problems with that it's unlivable. I'd rather just not have a television than continue to use OSMC.

In the meantime, I built a new gaming PC for my desk, the old machine (which happens to be in a Fractal Node 202 case so it already looks like a TiVo) has been moved into the living room. It's a Ryzen 3600/GeForce GTX-1080 machine with a bit over a terabyte of SSD and 16GB of RAM. It's still kicking bubblegum and chewing ass. Yes, it idles at a greater power draw than the Pi pulls at full steam, but it'll spend most of its time asleep, we'll be okay.

I'm currently still using Mint Cinnamon on it. Which is a hilariously unusable home theater OS. Plus I have my desktop (a Ryzen 7700/Radeon 7900GRE machine running Fedora KDE) and my phone (A Galaxy S10e) that I sometimes watch media on. Some questions:

• Why does VLC error out when trying to play mp4s stored on my NAS? Is it because SMB is Microsoft fucksewage that doesn't actually work? Because that's my working hypothesis.

• I have so little information on what Plex/Jellyfin even are, I gather Plex is at least semi-commercial while Jellyfin is the open source but worse option. These may or may not have a server component that has to run on a server-like box, which I don't and won't have.

• On the client side, I don't know if they take the place of a DE the way Kodi does, or if it's a separate app, or if you'd have to exit Plex or Jellyfin to use something like Steam, or if Steam Big Picture mode would work as a media center, but can it get to Youtube...

Is there anything out there that works better than throwing my TV away and forgetting about it?

AppFlowy is one of the open source Notion alternatives (a productivity tool)

New Features

• Made local AI free and integrated with Ollama
• Supported nested lists within callout and quote blocks
• Revamped the document's floating toolbar and added Turn Into
• Enabled custom icons in callout blocks

MediaWolf – A Missing Piece for the Arr Stack, Open for Contributors

Hey lemming self-hosters,

I came across MediaWolf recently and wanted to share it here. It’s an open-source project that’s tackling a big gap in the media automation space. If you use Radarr, Sonarr, Lidarr, Readarr, or any part of the Arr stack, you know how powerful they are for managing and downloading media. But despite how great they are individually, they don’t always work together as smoothly as they could.

That’s where MediaWolf comes in. It’s designed to tie everything together... acting as a discovery hub, recommendation engine, and management tool to help you find, organize, and automate your media library across multiple services. Instead of manually juggling requests between different apps, MediaWolf aims to make it all feel more seamless.

Why Does This Matter?

Right now, if you want to add a new movie or TV show, you probably:

Search for it manually across different Arr services

Check which quality profiles or indexers work best for that media type

Figure out if it's already available somewhere in your library

Manually add it to the right service

With MediaWolf, the idea is to bring everything under one roof... a central place to discover, recommend, and manage media without bouncing between multiple dashboards. Think of it as a missing bridge between the different pieces of your media automation setup.

The Project Needs Contributors

The creator, TheWickedWolf, recently posted on Reddit looking for developers to help bring MediaWolf to life. The project is already in motion, but for it to really thrive, it needs more people involved. Since it’s built in a modular way, you don’t have to commit to the entire project, just focus on a specific part that interests you.

New features? There’s room for innovation.

Bug fixes and improvements? Always welcome.

UI/UX enhancements? Could use some polish.

Open-Source and Accepting PRs

If you’re into media automation, open-source development, or just want to help improve how the Arr stack works together, this is a great project to check out. The team is actively accepting pull requests, and the creator is open to adding new contributors.

Repo: GitHub – MediaWolf

If you've ever wanted a smarter, more integrated way to work with the Arr stack, this might be it.

Given the recent news about Plex soon charging for remote access, I wanted to finish up my switch to Jellyfin.

What tools/methods have you all used to migrate watch history to Jellyfin?

I have a few family members in there, and would like to get everything switched over without resetting their watch history.

Dawarich is a Self-Hosted Location History Tracker (similar to the Google maps one)

Looks like this update includes improvements to the functionality for checking when you visited a place

I want a private place where I can talk to specific people.

I'd imagine I want something like:

• By default, nobody can register a new account on my server
• By default, nobody can view or join the rooms on my server
• If a friend has an account on a different matrix server, I can invite them to mine

I probably want some kind of federation with other instances (eg, where my friends might register their accounts), but not some free-for-all. Can someone recommend the right settings? The server is running synapse.

Hello there, I would like to upgrade my setup from a Raspberry PI 5 to something more capable of running Jellyfin properly. I also plan to host stuff like Nextcloud and some other small services.

The problem is I don't know anything about choosing hardware. I've tried reading online about this stuff and come across many recommendations to buy an NUC, not knowing I would have to purchase every component individually.

I would like your comments on the setup I currently have:

• NUC Type: ASUS NUC 14 Pro Kit (RNUC14RVKI300000I)
• Operating System - Still haven't decided yet, probably something easy like Ubuntu server
• CPU - Intel® Core™ 3 100U Processor
• Chipset - Integrated
• Graphics - Integrated - Intel® Graphics (C3)
• Memory - Adata 8GB DDR5 4800MHz CL40 SO-DIMM
• Storage - Western Digital Green SN350 1TB SSD M.2 2280 PCIe NVMe WDS100T3G0C
• LAN - Intel®, 2.5G LAN

This is a fork of the original project that is way ahead now.

cross-posted from: https://lemmy.world/post/27088416

This is an update to a previous post found at https://lemmy.world/post/27013201

---

Ollama uses the AMD ROCm library which works well with many AMD GPUs not listed as compatible by forcing an LLVM target.

The original Ollama documentation is wrong as the following can not be set for individual GPUs, only all or none, as shown at github.com/ollama/ollama/issues/8473

AMD GPU issue fix

1. Check your GPU is not already listed as compatibility at github.com/ollama/ollama/blob/main/docs/gpu.md#linux-support
2. Edit the Ollama service file. This uses the text editor set in the $SYSTEMD_EDITOR environment variable.

sudo systemctl edit ollama.service

3. Add the following, save and exit. You can try different versions as shown at github.com/ollama/ollama/blob/main/docs/gpu.md#overrides-on-linux

[Service]
Environment="HSA_OVERRIDE_GFX_VERSION=10.3.0"

4. Restart the Ollama service.

sudo systemctl restart ollama

We are also changing how remote playback works for streaming personal media (that is, playback when not on the same local network as the server). The reality is that we need more resources to continue putting forth the best personal media experience, and as a result, we will no longer offer remote playback as a free feature. This—alongside the new Plex Pass pricing—will help provide those resources. This change will apply to the future release of our new Plex experience for mobile and other platforms.

I am setting up nextcloud AIO in a podman container on my VPS. After some struggle, I got to the installation page, but domain checking is simply not working out.

After looking up, I decided to check the port from host machine. Strangely, curl localhost:11000 hangs indefinitely. nextcloud-aio-domaincheck container is running, and it mapped port as 0.0.0.0:11000->11000/tcp. The domaincheck server should be reachable, and I don't think firewall would be preventing localhost access.. The single line log from domaincheck container is:

2025-03-20 13:47:43: (../src/server.c.1939) server started (lighttpd/1.4.76)

I am utterly lost here. Does anyone know what would be possible reasons, and how to troubleshoot the issue? Any pointers would be greatly appreciated. Thank you in advance!

I've searched around and can't find this but it seems like someone must have created this already. I am hoping to find a self-hosted image resizer app. I frequently need to take photos from my phone (etc...) and make them small enough to post online.

For instance, my lemmy instance (lemm.ee) only allows images in posts if they are smaller than 500KB but my phone's photos are always larger than that.

In a perfect world, I could just browse to a local server app, upload an image, select a size to resize it to, hit Go and then download the smaller image. It doesn't have to allow any other editing and it shouldn't store images long-term. I want to self-host so I don't have to upload my images to random web sites I know nothing about.

I would be happy with a FOSS desktop app I can install in linux too, but then I couldn't access it from my phone. The Android apps I found for this either look scammy or include tons of ads.

Anyone know of such a thing? Thanks!

EDIT - UPDATE:

Thank you for the great suggestions. I've installed Image Toolbox on my Android phone and that looks great. It both has a ton of tools but also makes this resizing task very straightforward. Not sure how I never found that before.

But for my desktop, I started writing a PHP app to run in my existing nginx web server. It runs the suggested ffmpeg command under the hood, and since I am the only user on this server, this works very well for me.

That's working now so I am going to tweak a few things and then use it for a while. (Before anyone asks, I started based on the Python recs here, but couldn't get it working (PIP couldn't add Flask because PIP couldn't find PIP???) and so switched to PHP since my local server was already using that from another home-made app. This (PHP) was not as hard as I was afraid it would be, with help from Duck Duck Go's AI chat bot.

I used my php app to shrink this file!

Thank y'all - this is resolved now.

Following a suggestion i saw last day on a post here i have installed AList (https://alist.nn.ci/) and... guys it rocks!

It is file manager, both browser and WbDAV based. It can be configured with a ton of storages, so you can merge local and remote shares in the same place... You can use LDAP, OIDC and local authentication... WebDAV just works without hassle or fuss...

Moreover it's a chinese project and i got rolling over more than once trying to decypher the pseudo-english documentation.

Very easy to install too..

As usual, here is my wiki page: https://wiki.gardiol.org/doku.php?id=services%3Aalist

I’ve been experimenting with building and deploying ML and LLM projects for a while now, and honestly, it’s been a journey.

I had a really good conversation with Dean Pleban (CEO @ DAGsHub which is built on chromaDB and other OSS), who shared some great practical insights based on his own experience helping teams go from experiments to real-world production.

Sharing here what he shared with me, and what I experienced myself -

Data matters way more than I thought. Initially, I focused a lot on model architectures and less on the quality of my data pipelines. Production performance heavily depends on robust data handling—things like proper data versioning, monitoring, and governance can save you a lot of headaches. This becomes way more important when your toy-project becomes a collaborative project with others.

LLMs need their own rules. Working with large language models introduced challenges I wasn’t fully prepared for—like hallucinations, biases, and the resource demands. Dean suggested frameworks like RAES (Robustness, Alignment, Efficiency, Safety) to help tackle these issues, and it’s something I’m actively trying out now. He also mentioned “LLM as a judge” which seems to be a concept that is getting a lot of attention recently.

Some practical tips Dean shared with me:

Save chain of thought output (the output text in reasoning models) - you never know when you might need it. This sometimes require using the verbos parameter.

Log experiments thoroughly (parameters, hyper-parameters, models used, data-versioning…).

Start with a Jupyter notebook, but move to production-grade tooling (all tools mentioned in the guide bellow 👇🏻)

To help myself (and hopefully others) visualize and internalize these lessons, I created a guide that breaks down how successful ML/LLM projects are structured. If you’re curious, you can explore it here -

I MARKED the OSS ones!

https://www.readyforagents.com/resources/llm-projects-structure

Hi, I'm fairly new to the self-hosted universe but I like the idea of self-hosting media (I've looked at Jellyfin and Plex). But as I understand this requires quite some money and a lot of work. I don't think it's worth it if I put in all that effort just for myself but I'd love to build a small private streaming between me and my friends. We used to share and swap blu-rays after all, so it would be cool to build a shared collection.

My question is if that's possible and if anyone has experience with this? I've read that Jellyfin and Plex are meant as home-media-servers and I'm not sure what limitations that implies. Can people access the library from outside networks and will that affect the streaming quality/speed? What specs would the server need to ensure it can handle a bunch of users? Is there a software that is better suited for this use-case?

Thanks in advance for any help!

Discord was already succumbing to enshitification. Now with their intention to be owned by Wall Street, that trajectory will certainly accelerate at warp speed once the change of hands happens.

Anyone already get ahead of this and find a solid alternative?

Right now I'm on the fence between Element for Matrix, and Revolt. Both seem to have their pros and cons and I can't find a clear "winner".

cross-posted from: https://lemmy.world/post/27031457

CrowdSec "Community"

CrowdSec "Community" offering only gets worse and worse!

First, they had raised a paywall around querying details on IP addresses that triggered Alerts. Only 30 queries per week for the "Community".

Now, they have extended that paywall to cover the whole Alerts feature! Only 500 alerts per month for the "Community"!

Enshitification meets cybersecurity!

I am actually tried the different matrix variants. Conduit is my favourite. But I can't figure out how to setup a bridge with it.

I started with telegram. The appservice is registered and mautrix container logs are clean. I even can talk to the telegrambot but login isn't possible (since 2023) but what's the alternative? Really lost at this point.

What are you folks using for self-hosted single sign-on?

I have my little LDAP server (lldap is fan-fucking-tastic -- far easier to work with than OpenLDAP, which gave me nothing but heartburn). Some applications can be configured to work with it directly; several don't have LDAP account support. And, ultimately, it'd be nice to have SSO - having the same password everywhere if great, but having to sign in only once (per day or week, or whatever) would be even nicer.

There are several self-hosted Auth* projects; which is the simplest and easiest? I'd really just like a basic start-it-up, point it at my LDAP server, and go. Fine grained ACLs and RBAC support is nice and all, but simplicity is trump in my case. Configuring these systems is, IME, a complex process, with no small numbers of dials to turn.

A half dozen users, and probably only two groups: admin, and everyone else. I don't need fancy. OSS, of course. Is there any of these projects that fit that bill? It would seem to be a common use case for self-hosters, who don't need all the bells and whistles of enterprise-grade solutions.