Linux

I really wish that I was born early so I've could witness the early years of Linux. What was it like being there when a kernel was released that would power multiple OSes and, best of all, for free?

I want know about everything: software, hardware, games, early community, etc.

Hello!

I am pleased to announce a new version of my CLI text processing with GNU sed ebook. This book heavily leans on examples to present features one by one. In addition to sed commands and options, regular expressions are also discussed in detail.

Links

You can read the book online here: https://learnbyexample.github.io/learn_gnused/

Interactive TUI app for exercises: https://github.com/learnbyexample/TUI-apps/blob/main/SedExercises

Feedback

I would highly appreciate it if you'd let me know how you felt about this book. It could be anything from a simple thank you, pointing out a typo, mistakes in code snippets, which aspects of the book worked for you (or didn't!) and so on. Reader feedback is essential and especially so for self-published authors.

Happy learning :)

I am trying to update from Silverblue 41 to 42 (fully updated) but run into issues when attempting to update from both the software app and from CLI.

The problem using the software app is the same as what is described by this other user, who is using Fedora Workstation not Silverblue like I am:

https://discussion.fedoraproject.org/t/update-to-fedora-42-fails-in-gnome-software/148885

When I click the download button, it looks like it's downloading multiple files since the progress bar goes from 0 to 100 several times, and then it gets up to 95% then suddenly returns to the download button. This happens in about 30 seconds.

Using the CLI method, I run the following command:

rpm-ostree rebase fedora:fedora/42/x86_64/silverblue

and get the following errors:

 Problem: conflicting requests
  - package dnf5-plugin-automatic-5.2.12.0-2.fc42.x86_64 from updates requires libcurl-full(x86-64), but none of the providers can be installed
  - package dnf5-plugin-automatic-5.2.12.0-1.fc42.x86_64 from fedora requires libcurl-full(x86-64), but none of the providers can be installed
  - package dnf5-plugin-automatic-5.2.12.0-2.fc42.x86_64 from updates-archive requires libcurl-full(x86-64), but none of the providers can be installed
  - package libcurl-minimal-8.11.1-4.fc42.x86_64 from @System conflicts with libcurl(x86-64) provided by libcurl-8.11.1-4.fc42.x86_64 from fedora

SOLUTION: Uninstalled layered packages in dnf-automatic, libreoffice, and rpmfusion and then restarted. Rebase command successfully completed thereafter.

I want to create a USB gadget with a raspberry pi zero 2W. I'm starting with imitating a webcam I already have to see how much of this I can figure out. I've used the online documentation and a couple AI bots to get this far quickly, but I'm hung up on a ln command. It's telling me "ln: failed to create symbolic link 'configs/c.1/uvc.usb0': No such file or directory" when trying to create the link. This makes no sense to me though. I'm trying to create the link, of course it doesn't exist yet. That's what that command is supposed to do.

I've confirmed this problem in alpine linux and raspbian lite.

Below is the little script I have so far just to create the device:

#!/bin/bash
modprobe libcomposite
cd /sys/kernel/config/usb_gadget/
mkdir -p fauxcam
cd fauxcam
echo 0x046d > idVendor  # Logitech Vendor ID
echo 0x094b > idProduct # Brio 105 Product ID
echo 0x0200 > bcdUSB
echo 0x9914 > bcdDevice
mkdir -p strings/0x409
echo "111111111111" > strings/0x409/serialnumber
echo "Brio 105" > strings/0x409/product
mkdir -p configs/c.1/strings/0x409
echo "UVC Configuration" > configs/c.1/strings/0x409/configuration
echo 250 > configs/c.1/MaxPower
mkdir -p functions/uvc.usb0
ln -s functions/uvc.usb0 configs/c.1/
echo "usb0" > UDC

Back in August 2024, I discovered a linux distro called Dragora GNU/Linux. Dragora has a unique package manager called qi. For package instalation, Qi feel like installpkg from Slackware. For building package, qi takes some similiar approaches from PKGBUILD (Archlinux) and APKBUILD (Alpine linux). Qi uses a recipe as alternative of PKGBUILD, APKBUILD or slackbuild file to build package. Qi seems universal to use across distros, I’ve been actively using it on Ubuntu, Dragora, and LFS.

Hey again.

Thank you again for all of the help with the dual boot and repartition a few weeks back. I am running Linux Mint.

I repartitioned the Linux side to about 25 GB and over the last few weeks just downloading updates, I guess it has filled up. It tells me there is only 75 MB left. Is that normal or can I free up room again?

Also, the printer no longer prints. It just hangs when I try to print. It shows up correctly as the HP Deskjet 3510 but won’t print. Any tips on how to fix?

https://sw.kovidgoyal.net/kitty/kittens/panel/

Solution: When I formatted all my drives to install Linux on one and Windows on the other, I kept both connected and they share EFI boot partition as a result. Every time I reinstall Linux it formats the drive and therefore deletes the Windows's EFI Boot as well. One way is to fix this is to reinstall Windows while disconnecting the drive you have Linux on. Or you can move the boot files if you don't want to do that.

I used this guide: https://forums.tomshardware.com/threads/changing-windows-boot-manager-drive.3571420/post-21561626

Also remember to delete the Microsoft folder in the boot folder on Linux after you’ve checked that the new boot loader is working.

OP:

Currently dual booting as I need Windows for a few tasks and ganes Linux just won’t do. Since setting everything up I’ve reinstalled Linux twice, both times I’ve lost the ability to boot into windows and have needed to reinstall it.

Disk doesn’t show at all in Grub, tried all kinds of things but it just doesn’t show as a bootable OS. It doesn’t show in the boot options in the BIOS or the boot menu for my motherboard. Drive shows up and all the files are still on it. So my guess is the Windows bootloader somehow installs on the same disk that I have Linux on.

I run Linux(Fedora) and Windows on two separate drives.

Windows take forever to install. Anything I can do now to prevent this from happening if I need to reinstall Linux or if I wanna to some distro hopping?

Just to be clear, everything is working right now. But I want to prevent having to reinstall Windows every time I change distro or reinstall my Linux OS

Hello ladies and gentlemen as a brand new user (installed yesterday on new computer) of Linux in general and Bazzite specifically. I had a bootable USB I was going to use for a different distro before I decided on using Bazzite with another USB.

I decided to use the first one to move my meme collection to the new computer but when I deleted the partition and reallocated it with the highlighted option (the one that is not ms-dos I can not remember the specific name) the drive now seems to have disappeared. When I plug it in now it does not auto detect anything and for the life of me I can not find any drives through Dolphin.

If anyone can tell me how I fucked it up and/or how to find it/ fix it I would be grateful. I can always do it in Windows since I have to set up the old one to access the memes anyway but I would like to know how to do it here for the future. Thanks in advance.

EDIT: This has worked, thanks for help: LD_PRELOAD="" VK_LOADER_LAYERS_ENABLE=VK_LAYER_MANGOHUD_overlay_x86_64 %command% --skip-launcher --vulkan

Hi, so I've been using Fedora 41 GNOME since release with no issues at all and I've decided to do a new fresh install of Fedora 42 yesterday.

Everything seemed to run well but I've encountered this issue in games that after around 30min I get this weird stutter. Until then everything runs smoothly.

As you can see in the video the stutter only occurs during mouse movement or during camera movement with keyboard. Once the camera moves on it's own and just tracks the character the frametimes are perfectly flat so it does not seem like the fault is on the game but somethings off with the system compositor?

This happens with or without VSync, I've tried with and without VRR, I've tried chaning game settings and also different Proton versions... only thing that helps is to restart the game but then I'll have to do it once again in about 30min.

My suspicion is on the new triple buffering in new GNOME 48 but I have no idea how to turn it off to test.

Any suggestions?

Every few minutes, when I'm playing 3d games (like minecraft or blue price) my fps drops to around 5-10, stays that way for about 10 seconds, and then returns to a normal framerate. This does not happen with 2D games. I don't play competitive games so it doesn't impact me a ton but I thought I'd report it here in case someone can solve the issue.

Specs: Operating System: Fedora Linux 41 KDE Plasma Version: 6.3.4 KDE Frameworks Version: 6.13.0 Qt Version: 6.8.2 Kernel Version: 6.13.10-200.fc41.x86_64 (64-bit) Graphics Platform: Wayland Processors: 8 × 11th Gen Intel® Core™ i7-1165G7 @ 2.80GHz Memory: 15.3 GiB of RAM Graphics Processor: Intel® Iris® Xe Graphics Manufacturer: Dell Inc. Product Name: XPS 13 9305

My laptop does support this feature since it was working on Fedora KDE. But jumping over to arch, it seems not to work at all.

1. power-profiles-daemon.service is enabled and running.

● power-profiles-daemon.service - Power Profiles daemon
     Loaded: loaded (/usr/lib/systemd/system/power-profiles-daemon.service; enabled; preset: disabled)
     Active: active (running) since <time>; 12min ago
 Invocation: 4f20b3d144584a759b4a6c5ea14aa739
   Main PID: 608 (power-profiles-)
      Tasks: 4 (limit: 6850)
     Memory: 1.6M (peak: 2.8M)
        CPU: 81ms
     CGroup: /system.slice/power-profiles-daemon.service
             └─608 /usr/lib/power-profiles-daemon

Apr 18 11:14:52 berserk-arch systemd[1]: Starting Power Profiles daemon...
Apr 18 11:14:52 berserk-arch systemd[1]: Started Power Profiles daemon.

2. plasma-powerdevil.service is static and running.

● plasma-powerdevil.service - Powerdevil
     Loaded: loaded (/usr/lib/systemd/user/plasma-powerdevil.service; static)
     Active: active (running) since <time>; 12min ago
 Invocation: 7d72f24a0e5e4a74889a3895b91eb51c
   Main PID: 1074 (org_kde_powerde)
      Tasks: 9 (limit: 6850)
     Memory: 10.6M (peak: 11.4M)
        CPU: 1.391s
     CGroup: /user.slice/user-1000.slice/user@1000.service/background.slice/plasma-powerdevil.service
             └─1074 /usr/lib/org_kde_powerdevil

3. upower.service is enabled and running.

● upower.service - Daemon for power management
     Loaded: loaded (/usr/lib/systemd/system/upower.service; enabled; preset: disabled)
     Active: active (running) since <time>; 12min ago
 Invocation: 7aa43a43146346e383c961ce12cc9ded
       Docs: man:upowerd(8)
   Main PID: 540 (upowerd)
      Tasks: 4 (limit: 6850)
     Memory: 5.1M (peak: 5.9M)
        CPU: 251ms
     CGroup: /system.slice/upower.service
             └─540 /usr/lib/upowerd

I've already tried to to put

GRUB_CMDLINE_LINUX_DEFAULT="amd_pstate=active"

as a kernel argument that doesn't seem to do anything as well. I can't figure it out. The power management settings work tho. Any idea what's wrong? Thanks.

Title is quite self-explanatory, reason I wonder is because every now and then I think to myself "maybe distro X is good, maybe I should try it at some point", but then I think a bit more and realise it kind of doesn't make a difference - the only thing I feel kinda matters is rolling vs non-rolling release patterns.

My guiding principles when choosing distro are that I run arch on my desktop because it's what I'm used to (and AUR is nice to have), and Debian on servers because some people said it's good and I the non-rolling release gives me peace of mind that I don't have to update very often. But I could switch both of these out and I really don't think it would make a difference at all.

Hardware enablement highlights Canonical continues to enable Ubuntu across a broad range of hardware. The introduction of a new ARM64 Desktop ISO makes it easier for early adopters to install Ubuntu Desktop on ARM64 virtual machines and laptops.

Qualcomm Technologies is proud to collaborate with Canonical and is fully committed to enabling a seamless Ubuntu experience on devices powered by Snapdragon®

Ubuntu’s new ARM64 ISO paves the way for future Snapdragon enablement, enabling us to drive AI innovation and adoption together.

Leendert van Doorn, SVP, Engineering at Qualcomm Technologies, Inc

The Kubuntu Team is happy to announce that Kubuntu 25.04 has been released.

Codenamed “Plucky Puffin”, Kubuntu 25.04 continues our tradition of giving you Friendly Computing by integrating the latest and greatest open source technologies into a high-quality, easy-to-use Linux distribution.

The release features the latest KDE Plasma 6.3 desktop, KDE Gear 24.12.3, kernel 6.14, and many other updated applications and libraries.

EDIT: The bad solution is to unblock UDP port 5353 but the port has to be source port, not destination port. (--sport flag) See the now modified rules. The issue is that this is very insecure (see this stackexchange question and comments) but obviously better than no firewall at all because at least I'm blocking TCP traffic.

The proper solution (other than using glibc and installing nss-mdns package) is to open a port with netcat (nc) in the background (using &) and then listen with dig on that port using the -b flag.

port="42069"
nc -l -p "$port" > /dev/null || exit 1 &
dig somehostname.local @224.0.0.241 -p 5353 -b "0.0.0.0#${port}"

Then we need to remember to kill the background process. The DNS reply will now be sent to port 42069, so we can just open it with this iptables rule:

-A INPUT -p udp -m udp --dport 42069 -j ACCEPT

---->END OF EDIT.

I want to setup iptables firewall but if I do that, it blocks multicast DNS which I need. I am using command

dig "somehostname.local" @224.0.0.251 -p 5353

to get the IP through mDNS and these are my iptables rules (from superuser.com):

*filter

# drop forwarded traffic. you only need it of you are running a router
:FORWARD DROP [0:0]

# Accept all outgoing traffic
:OUTPUT ACCEPT [623107326:1392470726908]


# Block all incoming traffic, all protocols (tcp, udp, icmp, ...) everything.
# This is the base rule we can define exceptions from.
:INPUT DROP [11486:513044]

# do not block already running connections (important for outgoing)
-A INPUT -m state --state RELATED,ESTABLISHED -j ACCEPT

# do not block localhost
-A INPUT -i lo -j ACCEPT

# do not block icmp for ping and network diagnostics. Remove if you do not want this
# note that -p icmp has no effect on ipv6, so we need an extra ipv6 rule
-4 -A INPUT -p icmp -j ACCEPT
-6 -A INPUT -p ipv6-icmp -j ACCEPT

# allow some incoming ports for services that should be public available
# -A INPUT -p tcp -m tcp --dport 22 -j ACCEPT
# -A INPUT -p udp -m udp --dport 5353 -j ACCEPT # does not help
-A OUTPUT -p udp -m udp --sport 5353 -j ACCEPT # SOLVES THE ISSUE BUT IS INSECURE - not recommended


# commit changes
COMMIT

Any help is welcome :)

Hi all. This is an update on my script extracting CRC32 checksum from the 7z commandline tool. The output should be similar to how the md5sum tool outputs, the checksum and the file name/path.

The initial version of this script was actually broken. It would not output all files if a directory was included (wrong counting of files through argument number). Also filenames that contained a space would only output the first part until the space character. All of this rookie mistakes are solved. Plus there is a progress bar showing what files are processed at the moment, instead showing a blank screen until command is finished. This is useful if there are a lot of files or some big files to process.

Yes, I'm aware there are other ways to accomplish this task. I would be happy to see your solution too. And if you encounter a problem, please report.

crc32sum:

(Note: Beehaw does not like the "less than" character and breaks the post completley. So replace the line cat %%EOF with or copy it from the Github Gist link below:)

#!/usr/bin/env bash

if [[ "${#}" -eq 0 ]] || [[ "${1}" == '-h' ]]; then
    self="${0##*/}"
    cat %%EOF
usage: ${self} files...

Calculate CRC32 for each file.

positional arguments:
  file or dir       one or multiple file names or paths, if this is a directory
                    then traverse it recursively to find all files
EOF
    exit 0
fi

7z h -bsp2 -- "${@}" |
    \grep -v -E '^[ \t]+.*/' |
    \sed -n -e '/^-------- -------------  ------------$/,$p' |
    \sed '1d' |
    \grep --before-context "9999999" '^-------- -------------  ------------$' |
    \head -n -1 |
    \awk '$2=""; {print $0}'

Every four months, the KDE community rolls out a new wave of app releases all at once.

These updates cover a wide range of needs. Whether you’re managing personal files on your laptop or overseeing servers located thousands of miles away, KDE offers powerful tools to help you stay in control. Need to troubleshoot someone’s system remotely from the comfort of your sofa? There’s an app for that, too. From creating short viral clips for social media to producing full-length documentaries, KDE’s creative tools have you covered. And when it’s time to unwind, you can count on KDE for enjoying music, movies, or a good book.

Keep reading to discover what’s new in KDE Gear 25.04…

Upon upgrading from Fedora 41 to 42, I noticed that the USB tethering just doesn't work

networkctl recognizes the device as "wwan" now, instead of "ether". If I load up a previous kernel, USB tethering works normally.

This seems to be a change in the kernel, and probably won't be reverted in the future. What do I do?

More detail in this comment

You know what? Fuck this. I just backed my shit up and installed Debian. Still, please try to look into this, this could be a problem for many others

I'm leaving text editors like vscode/codium behind to learn something more modular, like Helix. I really wanna get used to. What advices can you give me to practice? I know that there is a :tutor command, I'm almost done with it. Do anyone know if there are exercices to practice? Im looking something similar to Ruby koans, a list of excersices to solve like "puzzles" but to Helix.

And also on computers generally lol.

The situation: I'm trying out Bitwig on my geriatric computer, which is running Linux Mint. It seems that I can't do very much without spiking the DSP, leading to awful glitchiness in playback. However, according to btop, the CPU (i7 4770) load isn't breaking 30%, spread evenly across the cores.

Things I have tried:

• uninstalling speech dispatcher, which helped
• tweaking the pipewire config, which doesn't seem to have helped much

So... what is the bottleneck here?

EDIT: the (main) issue was that my user didn't have real time priority permissions. An edit to /etc/security/limits.conf has improved things immeasurably.