In 2014, I bought my first network attached storage (NAS): the Synology DS 213j. Back then, I used it for storage-related tasks only, such as photo backup and media server. In 2020, I then bought the Synology DS 720+. At that time, I started experimenting with Docker and thought it would be great to have my own server that I could use for more than just storage-related tasks. This was the start of my self-hosted journey.
Synology DS 720+
The 720+ is a two-bay NAS with a quad-core processor and 2GB of RAM, which can be upgraded. It offers low power consumption, which was important for me to avoid fearing the electrical bill and to never feel the need to switch off the server. Having gathered experience with the DS 213j, I chose to stick with Synology. The applications they offer out of the box are easy to set up and keep running. Additionally, it is possible to run Docker containers on it, making it a versatile server.
Self-hosted Services
The first self-hosted service was the weather station I created along with an MQTT broker. The current services I host vary a lot. Whenever I stumble across a new project that can be self-hosted and looks interesting to me, I tend to give it a try. Depending on my liking, they will remain around or be stopped and deleted. Self-hosted services that have been around for a long time are Home Assistant, Paperless-ngx, and FreshRSS.
Why Self-hosted?
When you host services yourself, you are in control of your data. This makes migrating to a different service (e.g., switching your note-taking app) easier. It’s important to check how the data is stored; for note-taking, I prefer plain text, mostly Markdown.
A trigger for me to think more about my data was during the Covid pandemic when I was cycling a lot. A fitness watch tracked my data, and the provider visualized it well. However, retrieving the data from the provider’s server after selling the watch was difficult. Fortunately, several people on GitHub shared scripts to retrieve the data, often with the caveat, ”.. works until version X.Y.Z”. This was a wake-up call to take better care of my data.
This doesn’t mean I avoid all services, but I consider the accessibility of stored data more carefully. One question I often ponder: Why would a company offer free online storage, mail hosting, and a calendar? It’s not truly free.
However, using online services can be convenient and perfectly fine. It’s crucial to understand why something is provided to you and whether you can retrieve your data without issues later. You might be content with your current storage method, but that could change.
The Future
My own server started greatly with the Synology ecosystem. But lately, I also felt that the server could use some more computing power — especially if you want to use some LLMs. I am far from any choice, but the topic “Home Lab” is discussed a lot online, and there are plenty of YouTube channels giving great introductions to this rabbit hole. One thing to consider when moving away from Synology: Any service updates have to be done by me, and if something breaks, I also have to fix it.
Final Thoughts
Through my NAS, I had a great introduction to home servers and self-hosted services. It is a great starting point to consider your private data more thoroughly. With the rise of AI, it would also be great to have more computing power available. Use cases can be an automatic tag creation for documents added to Paperless-ngx. Since these documents can contain sensitive data, I would not want to use an online provided service.
Enjoyed the post? Have questions or feedback? I'd love to hear from you! Feel free to drop me an email at blog@jerey.at.