SpiderUnderUrBed

Currently I have only one laptop, I use it for everything and it has an nvidia gpu and a 11th Gen Intel(R) Core(TM) i7-11800H. I have a mode which disables the GPU which i use when i am out and about. However I want to minimize battery use to a significant degree, so which out of the three should I have on? power-profiles-daemon or autocpu-freq or tlp, which is better for battery performance? I know ppd is configurable via desktop, so i have kde set to power saving mode, but I still want better if possible, is there something more i can be doing with ppd or would autocpu-freq or tlp work better?

I am using rust, but this applies to many other languages, I get warnings like, dead code, unused variables, and etc, and while I remove most of them, there are some im not sure of, I like running my program and there being 0 warnings, or 0 warnings as i scroll down my code, so for things im unsure of, i mark them so the compiler doesn't put warnings there. I also add comments, starting with TODO:, it has some information about what to think about when i revisit it, also the todo's gets highlighed in my IDE with my extension, but is this bad practice?

I know this is application specific, but this relates to my overrall nix config i think, pipewire, wireplumber, all enabled. But I cannot record internal desktop audio, how do i fix this? is it a nix fix

I set the recording device to pipewire, I heard the host device needed to be set to pipewire or smthin, but only alsa shows up. If any logs are needed tell me.

EDIT: Not really solved but I found a script that worked for pw-record, now ideally I would like to fix it for audacity at some point, or just some other gui audio recorder

cat > ~/Music/record-desktop.sh << 'EOF'
#!/bin/bash
OUTPUT="${1:-$HOME/Music/Recordings/recording-$(date +%Y%m%d-%H%M%S).wav}"
pw-record --target 0 "$OUTPUT" &
PID=$!
sleep 0.5
pw-link bluez_output.30:50:75:16:40:A6:monitor_FL pw-record:input_FL
pw-link bluez_output.30:50:75:16:40:A6:monitor_FR pw-record:input_FR
echo "Recording to $OUTPUT (PID $PID) — press Enter to stop"
read
kill $PID
echo "Done! Saved to $OUTPUT"
EOF
chmod +x ~/Music/record-desktop.sh

flatpak run org.qbittorrent.qBittorrent
bwrap: Unable to open lock file /app/.ref: No such file or directory

I tried flatpak repair, with --user and --system, that did not fix it, and I did a system re-install, or rather reinstalled all the packages Nothing worked.

I am using nixos on unstable, recently I broke the nix part of my system, and was advised to do a re-install, I did a re-install which involved usb booting into nixos, and running a command to reinstall all the packages and rebuild in my nix directory, but afterwards, no flatpak is running. There is a possibility it broke beforehand and I didnt notice that, but i think it might have been after the re-install.

Flatpak version: 1.16.1

Distro: NixOS

Distro version: /nix/store/8hx8jwa24q6rkzhca9iqy5ckk9rbphi1-source

Architecture: x86_64

EDIT: Some flatpaks run but alot of them dont with the same error

EDIT (x2) SOLVED: I think the issue might have been the kde platform, as the apps that crashed were on kde platform 6.8 or 6.9, possibly some broken version, most likely the platform and bwrap didnt play well with nixos, these commands worked, I ran this for users (not system). Possible system might work, point is, I dont have the issue anymore.

flatpak uninstall org.inkscape.Inkscape org.qbittorrent.qBittorrent com.obsproject.Studio org.kde.kdenlive org.wireshark.Wireshark com.prusa3d.PrusaSlicer com.rafaelmardojai.Blanket io.github.seadve.Mousai net.codelogistics.clicker org.freefilesync.FreeFileSync

flatpak uninstall --runtime org.kde.Platform//6.8
flatpak uninstall --runtime org.kde.Platform//6.9
flatpak uninstall --runtime org.gnome.Platform//48

flatpak uninstall --unused -y

flatpak repair --system
flatpak repair --user

flatpak install flathub org.inkscape.Inkscape
flatpak install flathub org.qbittorrent.qBittorrent
flatpak install flathub com.obsproject.Studio

This was talked about before, but the settings location that was mentioned to enable thumbnails on remote files and stuff has changed. I have a mount of my android system over mtp, and I cannot see any of the thumbnails, making it impossible for me to sort my stuff, some into hard drives and etc.

How do i enable seeing mtp thumbnails in kde, is there issues with just mtp thumbnails in dolphin or what alternative file manager or image veiwer I can use so i can see the thumbnails and relocate the images.

Hi, I want a nice easy-to-use graphical program which can show image thumbnails atleast, and provide an easy way to encrypt images, and ideally other files too. I heard of a few programs but some dont show images and some look a bit dated ui wise. Seeing image thumbnails is pretty much a requirement as it would be alot of swapping back and forth otherwise.

I am using KDE (and nixos), I dont think that matters much for what I want.

I heard varying things, I expect to have like 6-8 people come over in a few weeks and I want to set up lancache.net, but performance wise would it be horrible on a raspberrypi 4? or decent? I heard many things about lancache support on the raspberrypi, so I would like to hear about that too.

I do have a intelnuc as a secondary server but it would be harder to set up lancache on that, also I think I need to get a wifi card to set up a AP network, so would it make significant difference for a 16gb ram and modernish intel cpu to run it?

Here is a volume I created with two nodes, not replicated but rather distributed:

[nix-shell:~/hdd]$ sudo gluster volume info gv0
 
Volume Name: gv0
Type: Distribute
Volume ID: 8db02f86-0d22-48d5-952e-192b5eb4397c
Status: Started
Snapshot Count: 0
Number of Bricks: 2
Transport-type: tcp
Bricks:
Brick1: hostnuc:/home/spiderunderurbed/hdd/gluster/brick2
Brick2: raspberrypi:/home/spiderunderurbed/hdd/gluster/brick2
Options Reconfigured:
storage.fips-mode-rchecksum: on
transport.address-family: inet
nfs.disable: on

[nix-shell:~/hdd]$ 

Respective partitions both the bricks are on: Rpi:

└─sdc2
       8:34   0   1.9T  0 part /home/spiderunderurbed/hdd

Hostnuc:

├─sda1
│      8:1    0   1.6T  0 part /home/spiderunderurbed/hdd

When i run df -h where i mounted it:

spiderunderurbed@raspberrypi:~/gluster-mount $ df -h .
Filesystem      Size  Used Avail Use% Mounted on
localhost:/gv0  1.9T   20G  1.9T   2% /home/spiderunderurbed/gluster-mount
spiderunderurbed@raspberrypi:~/gluster-mount $ 

I only see the size of my pi's partition, and not the full 3.5T i should be getting, why is this? and how to fix it.

Ncdu takes ages to run on my system, its like 500GB+ of storage space, but it takes roughly an hour to finish scanning, probably a bit less, is there any alternative which either constantly monitors my files so it always knows the sizes for me to navigate or is significantly faster than Ncdu?

Is there any js module for lazy-loading a files content? in vanilla js, (that a import via a script tag), so that if i scroll enough, it will send a request to the server for more of a files content, at a specific chunk of it, and the browser will add the rest of the content to the file viewer/editor

I am making a filebrowser, and so far i implemented navigating folders, uploading files, next I would like to make a file editor, so its not the html page itself I want lazy-loaded, I want a file to be lazy-loaded into the editor, again, so you scroll enough, and it will fetch the chunks from the server (i can figure out the backend bit, I would like it if it worked with multexjs or multipart in rust and I dont have to manually sort out file chunking but thats not a hard requirement.)

Alternative to modules would just be code i can use

The image shows what happens when i boot on the ISO, i get the error:

x sysroot-nix-rrv\v2dstore.mount - /sysroot/nix/.rc-store
    Loaded: loaded (/nix/store/1779scwag89fja51b1q7jg5savcj5tyrlnltrd-fstab; generated)
    Active: failed (Result: exit-code) since Tue 2023-07-29 18:37:53 UTC; 35s ago
Invocation: 94ea87636cc24a4c83d8e147275a8124
    Where: /sysroot/nix/.rc-store
    What: /iso/nix-store.squashfs
    Docs: man:fstab(f)
    man:systemd-fstab-generator(8)
    Item peak: 1.9M
    CPU: 9ms
Jul 29 19:37:53 hostruc systemdf11: Mounting /sysroot/nix/.rc-store...
Jul 29 19:37:53 hostruc mountt2531: mount: /sysroot/nix/.rc-store: failed to setup loop device for /iso/nix-store.squashfs.
Jul 29 19:37:53 hostruc systemdf11: sysroot-nix-rrv\v2dstore.mount: Mount process exited, code-exited, status=82/n/a
Jul 29 19:37:53 hostruc systemdf11: sysroot-nix-rrv\v2dstore.mount: Failed with result: exit-code .
Jul 29 19:37:53 hostruc systemdf11: Failed to mount /sysroot/nix/.rc-store.

and I am pretty confident this is because I have initrd systemd on a iso build, so does anyone know how I can fix this issue in either my flake where i set how the iso is built or in main.nix (which i linked), which is about initrd settings main.nix (imported in configuration.nix): https://pastebin.com/3VVv4epH flake.nix: https://pastebin.com/yzVrJ0YJ

is /run shared to the real filesystem after the initrd stage? like what directories are shared to the real filesystem after initrd, or what files. Or is everything just inaccessible in the real filesystem from the initrd phase.