I've got an x86 tablet, and I don't want to run Windows or Android, so I have always been left with the option of Linux with GNOME. GNOME is a solid DE, but despite the design language, it is nowhere near ready for touchscreens. Depending on the distro, it was either severely outdated or just generally buggy, particularly with the onscreen keyboard. This lead my to completely abandon my tablet for a while, but today I had an arbitrary though that I should try out KDE Plasma Mobile. Low and behold, it was exactly what I needed for my tablet.

I expected it to be a bit buggy, because I had tried out KDE Bigscreen and this seemed like the mobile equivalent of bigscreen, a fork that is half-baked and still has lots of work to do before it's worth daily driving. Imagine my shock when I booted the Fedora liveUSB and everything felt incredibly fluent and polished. I knew about 15 seconds in that I was switching. The navigation feels like an Android tablet, which I see as a positive, and the keyboard is just so much larger and more responsive. The GNOME default OSK is hardly usable because it's letters are so small, so this was a relief being able to type with no need for checking that I'm writing correctly every few characters. The next thing I noted is how polished it feels even for large screens(I have a 12 inch tablet). The gestures work on the whole screen, and this makes it feel effortless to navigate around. The final thing I want to touch on before I stop glazing is the really good performance. I have a very old tablet, about 10 years old, and the performance felt like it was brand new.

Thanks for reading my glazing exhortation, and if you have a tablet, I couldn't recommend Plasma Mobile more.

P.S. does anyone know where KDE connect is?

A Qnap NAS has a drive with some bad sectors, I want to RMA it, but before just want to figure out how to prepare a drive? It's part of a raid 5 setup of 4 drives unencrypted. So I want to remove it and wipe it. Got a Linux machine I can use, but never done this before.

What are common Linux tools to do that sensibly?

Let's say I want to enforce certain settings, such as the use of a proxy in network settings for certain users.

Isn't this easily bypassable by for example by installing TOR browser or using a VPN app in the user space?

How does system mangers can be sure users will only use the system as planned by the sysadmin? I'm especially interested in network settings, but in general I would be interested to know more about this/be pointed towards the right direction.

Thank you!

Hello :)

There isn't any community about note taking where I could post my question and no this is not a "What's the best note taking app" question...

I'm getting tired of maintaining my Obsidian vaults... Somehow I'm fighting to get it right and obsidian seems to fight back. I've got 4 vaults of the same subject and I always end to make a mess out of it and make a fresh one... Also my notes a scattered in all direction and the more my knowledge base grows the less I seem to be able to find something...

This is probably a me problem rather than Obsidian issue. The way I'm taking notes are not compatible with Obsidian. IMO Obsidian's defaults configuration are bad and visually not appealing. Sure customization in Obsidian is "endless" but digging in the HTML code to change the style or adding plugins to somehow get something visually appealing seems more like a chore than actually taking notes.

Here I'm again roaming the web for a Note taking app the could fit my needs and after trying a lot of different apps (please don't suggest the already well known apps... I have probably already tried it...) I couldn't find something that fits my workflow.

The only one that looked great and simple was osmosnote but it isn't maintained anymore. There's also dendron but it's in maintenance mode. So there goes the only ones that looked promising from my perspective.

After giving it more thoughts, I was looking for something that could:

• Keep my scripts updated
• Simple markdown text
• No database
• Local first
• Open source
• If webapp self-hostable
• Back-linking
• Keep track of changes

Except for back-linking, a self-hosted Forgejo with git seems to fit all my needs, however I'm not sure if this is the right tool and I'm scared that in the long run I will mess it up the same way I did with Obsidian.

Does anyone here has some experience and is taking notes that way? I'm really curious on your experience and maybe your thoughts if it's feasible ? Practical ?

Please don't suggest Org.mode or Emacs ! They look very cool and very promising but they are WAY to much overkill ! And they also implement a totally new way of taking notes... Relearning on how to take notes will probably give me the last hit on abandoning to document anything !

Thank you for any helpful input !

The card in question: Thames [Radeon HD 7550M/7570M/7650M] (Radeon HD 7570M) I am starting to face mounting issues as a lot of modern software have started to abandon OpenGL in favor of Vulkan, and if its possible, I just want to try for the learning opportunity

Hardware Information:

• Model: HP Victus by HP Gaming Laptop 16-s1xxx
• Memory: 16.0 GiB
• Processor: AMD Ryzen™ 5 8645HS w/ Radeon™ 760M Graphics × 12
• Graphics: AMD Radeon™ Graphics, NVIDIA GeForce RTX™ 4050 Laptop GPU
• Disk Capacity: 512.1 GB

Software Information:

• Firmware Version: F.13
• OS Name: Fedora Linux 41 (Workstation Edition)
• OS Build: (null)
• OS Type: 64-bit
• GNOME Version: 47
• Windowing System: Wayland
• Kernel Version: Linux 6.12.9-200.fc41.x86_64

Issue Description:

I've been experiencing persistent issues with brightness and display on my Fedora workstation. Sometimes, the brightness becomes very low, and I can't fix it without rebooting the laptop. This issue also occurred on Ubuntu, and it happened twice recently.

Just a few minutes ago, my display went completely black, and I had to hold the power button to shut down and restart the laptop. Here are the relevant errors from the log:

nvidia 0000:01:00.0: [drm] Cannot find any crtc or sizes

...

ACPI BIOS Error (bug): Could not resolve symbol [_SB.PEP.DETY], AE_NOT_FOUND (20240827/psargs-332)

...

ACPI Error: Aborting method _SB.WMID.WMAA due to previous error (AE_NOT_FOUND) (20240827/psparse-529)

These issues don't occur consistently, making it difficult to reproduce them. I can perform all tasks normally, including gaming, without encountering these problems most of the time.

If I cannot resolve this issue once and for all, I may revert to Windows 11. Any suggestions or solutions would be greatly appreciated.

I know this may sound like an over the top useless tinkerink, but I just like to tinker with, play around with stuff and learn, and that may be why I use Linux on my notebook.

I have read some articles about ZRAM and what might be the best configuration and even chatted with some generative models, but didn't come to a decisive conclusion regarding size, compression algorythm etc... I am asking anyone interested to respond about their experience and recommendations given my specs:

OS: Fedora Linux 41 (Workstation Edition) x86_64
CPU: AMD Ryzen 3 7330U (8) @ 4.39 GHz
GPU: Amd Barcelo [Integrated]
Memory: 5.63 GiB (2 GiB is reserved by the GPU)

I use the notebook for school, mainly lightweight programming and using the browser.

I don't expect any magical improvements from the ZRAM as I already use it with the default config, just wanted to learn something new.

Change log for January 17, 2025 Vulkan 1.4.305 spec update: Github Issues

• Clarify dynamic array layer access in common draw dispatch VU 09600 and the <<resources-image-layouts, Image Layouts>> introduction (public issue 2474).
• Remove Ash CI in both Github and Gitlab (public PR 2479).

Internal Issues

• Consolidate common draw discard rectangle VUs and clarify behavior (internal issue 3400).
• Clarify the maxMemoryAllocationSize limit (internal issue 4119).
• Restrict vkCmdExecuteCommands VU 09376 to secondary command buffers (internal issue 4126).
• Fix AV1 decode frameHeaderOffset typo in vkCmdDecodeVideoKHR VU 09251 (internal MR 7043).
• Add more detail about VkAccelerationStructureCreateInfoKHR::deviceAddress (internal MR 7067).
• Clarify ambiguities related to <<decode-av1-picture-info, AV1 decode parameters>> FeatureEnabled, FeatureData, loop_filter_ref_deltas, and loop_filter_mode_deltas (internal MR 7070).
• Remove common buffer memory barrier and image memory barrier VUs already converted by VUIDs 10387 and 10388 (internal MR 7075).
• Remove redundant VkShaderCreateInfoEXT VUs 08452 and 08458 (internal MR 7076).
• Fix typo in VkSwapchainPresentScalingCreateInfoEXT VU 10154 (internal MR 7077).
• Remove redundant explicit VkPhysicalDeviceSurfaceCapabilitiesKHR VU 06523 (internal MR 7078).
• Remove redundant VkWriteDescriptorSetAccelerationStructureKHR and VkWriteDescriptorSetAccelerationStructureNV VUs 02236 and 03747 (internal MR 7079).
• Remove redundant vkCmdBindShadersEXT feature bit VUs 08474, 08475, 08490, and 08491, (internal MR 7083).
• Revert vkCmdWaitEvents2 dependencyFlags VU that should not have been removed (internal MR 7089).
• Reword common draw vertex binding vertex attribute VU 10390 (internal MR 7091).
• Fix asciidoctor markup for the VK_QCOM_image_processing proposal and update some descriptions (internal MR 7097).

https://flathub.org/

I was recently intrigued to learn that only half of the respondents to a survey said that they used disk encryption. Android, iOS, macOS, and Windows have been increasingly using encryption by default. On the other hand, while most Linux installers I've encountered include the option to encrypt, it is not selected by default.

Whether it's a test bench, beater laptop, NAS, or daily driver, I encrypt for peace of mind. Whatever I end up doing on my machines, I can be pretty confident my data won't end up in the wrong hands if the drive is stolen or lost and can be erased by simply overwriting the LUKS header. Recovering from an unbootable state or copying files out from an encrypted boot drive only takes a couple more commands compared to an unencrypted setup.

But that's just me and I'm curious to hear what other reasons to encrypt or not to encrypt are out there.

I am trying to restrict access to our workgroup shares so no one can accidentally delete or move them.

Basically /path/to/parent/SHARE WITH TEAM FOLDERS/

The directory in caps is all the different team shares in the company. Hr, sales, etc. the issue we have is everyone has access to the directory with all the shares and from there it’s governed by groups on the acl.

Sometimes users accidentally drag a folder into a different teams folder or deletes a team folder.

I can’t find a way to restrict access so that the groups applied can still have access but restrict delete or moves.

Any help would be greatly appreciated….i might actually cry if something works too.

Hi,

I would like to pass the arguments to apt trough a .sh script.

apt install "${1}/opensnitch_${1}_amd64.deb" "${1}/python3-opensnitch-ui_${1}_all.deb"

so for example if I launch test.sh 1.6.5 the script should run apt install 1.6.5/opensnitch_1.6.5_amd64.deb 1.6.5/python3-opensnitch-ui_1.6.5_all.deb

but it's not wokring it pass to apt only 1.6.5 and not the rest of the string... any ideas ?

SOLVED

Edit: Solved using workaround of creating a systemd service that disables GPP0 wakeup: https://forums.linuxmint.com/viewtopic.php?t=364220&start=20

Hey everyone!

The suspend feature has never worked on my system, so I always turned it off in the settings. But now, with the electricity prices going up in my country (we have the most expensive electricity in Europe, lol), I figured it’s a good time to tackle this.

When the system tries to go into sleep mode, it just wakes up right away, and the monitors don’t show anything (no signal). I’ve got a backlit keyboard and an RGB mouse, but they don’t light up during this, and the caps lock key doesn’t respond when I press it.

I’m on kernel 6.12.7 and I updated the BIOS a few days ago, but the problem is still there.

My hardware:

• CPU: Ryzen 5 5600
• MOBO: GIGABYTE B550 GAMING X V2 (rev. 1.3)
• GPU: RX 6700 XT
• PSU: SilentiumPC Supremo M2 (600W I guess?)
• CASE: be quiet! Pure Base DX500

If anyone has any ideas on how to fix this, I would really appreciate your help!

cross-posted from: https://programming.dev/post/24130558

My Win10 work laptop has a network share of a remote windows server. I access it everyday. If i change passwords, i have to remap the share.

I have a linux vm that does the builds for my project. It too has a mounted directory mapped to that remote windows share, using my credentials.

I tried mapping the share in another linux vm but got errors so ended up quitting as it wasn't that important.

However, now i can't access said share in any device, by name or IP address. WTF happened?

The mount command i use in linux is mount -t cifs -o rw,relatime,vers=default,cache=strict,username=my.username,domain=,uid=118,noforceuid,gid=130,noforcegid,addr=10.10.10.10,file_mode=0755,dir_mode=0755,soft,nounix,serverino,mapposix,rsize=1048576,wsize=1048576,echo_interval=60,actimeo=1 //10.10.10.10/dir1/dir2 /media/remoteshare, the UID/GID are of the user that runs the builds.

I'd get having errors on mounting the remote share, but i'd expect that to be limited to the local computer i was trying to mount on, not that it would propagate to any device that has this share mapped!

Up till now I've been using only OpenVPN connections.
Which can be integrated nicely in NetworkManager to connect + disconnect from them, through NetworkManager's UI.

However since recently I also need to use Netbird for some connections.

Is it possible to add Netbird as an entry in NetworkManager, for easy connection + disconnection?
If so, how would one go about setting that up?

Or is adding Netbird in NetworkManager not possible,
and is the Netbird CLI/GUI really the only option?

Edit:
It's not possible yet.
First Netbird needs to add Wireguard config support,
see: https://github.com/netbirdio/netbird/issues/2504

Hi,

On my Debian 12 Bookworm ( raspberry pi )

My nftables do not block DHCP packets ! according to this basic rules

nft -y list rulesset

table ip Tip {

	chain chIN {
		type filter hook input priority 0; policy drop;
		ct state established,related accept
	}
	chain chFW {
		type filter hook forward priority 0; policy drop;
	}
	chain chOUT {
		type filter hook output priority 0; policy drop;
		udp dport 67 drop #DHCP
		udp dport 53 accept
		tcp dport { 80, 443} accept
		ct state established,related accept
	}

}

DHCP should be blocked... but it's not as I get an IP from it..

Any ideas ?

Thanks.