

(I’m not very good with the command line)
Me either so I take a lot of notes about commands and command sequences. Also, I find that Grok is pretty decent at explaining commands. AI is a wonderful tool, but you also need to do your due diligence in ascertaining whether AI has given you the correct information. I would not copy/paste random commands into a production server, but rather I have a small test server for that kind of stuff. Once I have the command, tested, and understand the command, I can then use that in a production environment.
In as much as I love a good WUI, you will have to learn some cli, it’s just inevitable, especially in a headless environment. It may seem daunting at first because there are literally thousands of commands and command sequences. I honestly doubt if even the geekiest nerds on the planet know all by heart. For each command sequence, there are probably hundreds of ways to compose the same command. I would admonish you to download Notepad ++ and start keeping notes on the commands you use. Later on, the fun part is looking back on your notes to see all the commands you now know and what they do.
Core utilities (like ls, cd, cat, etc.) from projects like GNU Coreutils provide around 100-200 commands. Additional tools from packages (e.g., grep, awk, sed) and system utilities (e.g., systemctl, iptables) can add hundreds more. On a system with many packages installed, running compgen -c | sort -u | wc -l in a Bash shell might show 2,000–10,000+ unique commands, depending on the setup.
I’ve often thought about this, and since it has come up in convo, I’ll ask: If you were to implement a backdoor to your server, how would you go about that? Currently I have 3 vps and one rack in the closet. It is the vps I’m interested in the most. Only one vps offers a rescue ssh, and yes I can confirm, if you are not exceedingly careful on my setup, you can lock yourself right out. I run tailscale on everything and I often wondered if I could incorporate tailscale as a emergency backdoor.