Ever had a question about Linux but felt too afraid to ask? Well now’s your chance, ask any question about Linux, no matter how noob or repeated it is, and I and others will help answer them.
Previous noob question thread: https://lemmy.ml/post/14261893
What is something Linux related that you’ve learned recently?
As a meta question, could this work as an additional (or alternate) recurring discussion question? It felt similar in intent, to encourage people to keep learning / asking questions and chances are that if someone learned something then others will benefit from the information (or correct them)
After 26 years of using Linux, I did my first baremetal “immutable” distro install last week.
My youngest son is starting school and instead of the Chromebooks that they recommend, I took a chance and installed Fedora Silverblue on a $200 Lenovo “student-rugged” class laptop. Everything works and he hasn’t had any issues so far. He gets access to the same student platform as the other students through Chrome, but then I can install Minetest and Tux Paint and GCompris as well.
The older kids run Debian stable for years now, but if this works out, I might transition them over next semester.
I learned how a kernel actually loads a program and switches between them by using timer interrupts and interrupt vectors that point to specific locations in memory to resume execution from. Not specifically Linux related, but I’m trying to learn more computer science, and it just clicked for me two weeks ago. I’ve been programming microcontrollers for ten years, but those are monolithic programs, and while I knew what interrupts were and have used them, I never understood how an OS actually runs multiple things while staying in control. Now I do. About time I understood a core concept of these machines that have been here all 42 years of my life.
It’s one of those “aha!” moments like when I realized classes and structs are just data types like any other in C++ when I was starting off programming and can be used like them. OOP became fun after that.
I remember when the mapping of virtual memory segments clicked for me. I think i said out loud, “that’s so clever!”. Now it just seems so fundamental to managing memory for user space applications, but I hadn’t thought about how it was done before.
The other day I learned that you can just grep an unmounted filesystem device. It will read the entire disk sequentially like it’s one huuuuge file. And it will reveal everything on that disk… whether a file inode points to it or not.
Used it to recover data from a file I accidentally clobbered with an errant mv command. It’s not reliable, but when you delete a file, it’s usually not truly gone yet… With a little luck, as long as you know a unique snippet that was in it, you can find it again before the space gets something else written there. Don’t even need special recovery tools to do it, just use dd in a for loop to read the disc in chunks that fit in RAM, and grep -a for your data.