That sounds like an interesting idea, I'll test that out, thanks!
I thought about this solution, as it is the "cleanest", however I need on total 4 firefox derivatives. Unfortunately, when looking deeply into the options, i haven't found 4 that are similarly trustworthy, well maintained etc. Also i have my firefox config fully figured out, it works and is as private as i want them, without some maintainer forcing their opinion on my use cases. Plain firefox is the easiest to configure, as it's like a blank start. However i might be wrong here and am open to suggestions :D
Unfortunately that is not what I am looking for. I am already using named profiles. Like i stated in my original post as well as my answers below, this only works from Inside Firefox, however from the operating system pov it is still treated as the same application. Which means:
a) When i share the work profile, i also share all other profiles, as they are all Firefox b) When I quick access firefox via spotlight, i end up at the nearest, random profile / instance of firefox. c) There is no way to differentiate the profiles on an application level. d) I can not assign the instances to different desktops, as they are all Firefox.
Hello!
Quick question: How do i completely separate (also with distinct names) Firefox instances?
What i mean by that:
On my old Intel Mac, i've made 4 copies of Firefox, one for work, safe browsing, and apps. I've renamed each with FF-name so i can quickly open them with spotlight. Each has it's own profile.
For that i "only" need to edit the Info.plist and all worked.
However since then i migrated to Apple Silicon and MacOs Ventura, which seems to interfere with my approach.
Do you guys have an alternative idea or better approach? Maybe docker containers or anything?
My main intent is that i have them as separate applications, so i can open them with spotlight and also do not accidentally share the wrong instance in a conference call.
I had a similar idea, however i haven't seen a markdown plugin, that is well maintained and at the same time simple enough, so that the core, in this case markdown, can easily be replaced with a completely different engine, asciidoctor.
Any recommendations for that?
I also thought about changing neorg, but the missing support for treesitter is a k.o. for asciidoc.
The conversion is not an issue, there are already multiple tools for that, including a browser plugin with auto refresh.
However the tight integration with the editor, in this case neovim, is missing. At the bare minimum it should show the changed area curently being edited, ideally scroll with the editor scrolling like with common markdown extensions. Currently it just shows a static site that refreshs.
Hello Everyone,
i'd like to ask for your opinion on the following issue:
i've created my own knowledge base based on asciidoc, with some custom shell scripts and a go application for creating backlinks, tags etc. I've chosen this way, as most solitions are based on markdown, which is not standarized and very limited compared to asciidoc, especially from a dev pov.
All my editing, searching etc. is done via neovim, which is very comfortable.
However, i'd like to improve the user experience with the setup, as i'd also like to see the rendered version, especially when adding mermaid diagrams and other things.
I've tried some plugins for the browser, which render the view and update automatically, however they are not in sync with my nvim, so i have to scroll on every save, if i want to see the rendered version. That's not ideal.
Any ideas?
Ideally I'd like some kind of application template, where i can embed a terminal / neovim and a webbrowser, ideally linked via lua scripting, so it integrates nicely. It can also be a completely separate application like anytype, however i've not seen anything that has a proper vim-like module editing support & allows for asciidoc rendering instead of markdown.
Good choice, however the number of supported languages is limited to things with language server or treesitter support. Meaning languages without this, like asciidoc, is not supported and might never be
I recommend to start with kickstart if you prefer to customize https://github.com/nvim-lua/kickstart.nvim or Lunarvim for a full IDE experience to see what's possible and go from there https://www.lunarvim.org/de/
Most works out of the box then. Start there and customize further to your needs
It looks very interesting!
But I don't see the unique selling point of it compared to alacritty and kitty, besides web-enabled. Is there anything that it does better than these 2?
That looks promising, especially since my current status bar is also just a collection of shell scripts, so that might be easier to switch
Thanks! That looks exactly like what I was looking for. I hope it works as promising as it looks :)
Thanks, that was a very interesting read!
I forgot one essential tool, where I need a recommendation for: spotlight. I use it to switch quickly between applications or to folders. Keyboard shortcut, first letter of the application name and enter... I know there are solutions, but I only heard from Ubuntu, which I don't want. Anything simple and fast you can recommend?
Thank you everyone for all your suggestions! I'll quickly try to summarize them for myself. So what you suggest is:
Operating Systems:
- NixOs
- Debian 12
- ElementaryOS
- mint
- PopOs
- EndevourOS
- Fedora
- arch
- Opensuse
- Novara
Tiling Window Manager:
Recomended to use something based on wayland.
- hyprland (can be configured from file, good compatibility with nix)
- sway (proposed with Debian, multiple suggestions, config via file as well, good for custom keybindings, already options for sway in nixos)
- i3
- bspwm
- KDE Plasma
- dwm / dwl
Status Bar:
- swaybar (in case of using sway)
- waybar (when using wayland)
- eww
- ags
- KDE neon
Package Managers:
- flatpack
- brew (is this already stable enough?)
- Nix (obvious choice if nix os chosen)
- snap
- (pacman if arch)
- integrated one
Packages:
- together with wayland alacritty or kitty
- foot
- Yakuake
- suckless
At the moment I am trying to avoid anything where RedHat is involved. Not because of the recent controversy, but simply IBM is known to kill their software solutions on a whim. (although i still use ansible), so Fedora is unfortunately out (again, no judging on how great it is). I've been quite interested in EndevourOS, so that might be fun to try out. Debian for the desktop probably not right now. I'm running it on servers for stability, but for a desktop environment, i prefer having more recent packages (e.g. neovim). The "sales pitch" for Mint sounded pretty interesting as well. However i'll give NixOs a try first, simply because it was mentioned very often, same with sway.
Based on this i'll try out these combinations first:
- NixOs with sway and eww
- NixOs with hyprland and waybar
- NixOs with dwl and ?
If this does not satisfy, i'll look into endevourOS and mint, but that might require some Ansible I assume.
Thank you very much!
It might have. I've tried nixos on a mini PC meant as a home server, so most configuration is done via SSH and users don't change (much), I might have accidently activate it while trying nixos out.
Making users unable to login is a bit of an odd (side?) Effect, but maybe I'm not understanding the purpose of this option correctly. I'll stay away from it for now :D
Hello Everyone,
as you can see on my screenshot, i am using an intel based mac for years now, which i customized to my needs. However i have reached the limits of this machine in terms of customization options and would like to move to linux to test it out as a daily driver. I'm actually quite happy with mac from the pov that everything just works, however there are certain things that annoy me, but apple does not allow me to change them.
As a newbie in terms of desktop linux (i've used ubuntu roughly 12 years ago as a daily driver and am familar with headless linux), i'd like your advice.
Specifically I am looking for:
- a minimal, fast system
- keyboard / shortcut based - all interactions can be done from keyboard (within common sense limits)
- all keys can be custom mapped (i have muscle memory of my custom keys for certain actions, so i'd like to keep them)
- all can be configured from dotfiles (worse case shell scripts and ansible)
- very low ressource consumption, snappy system with no delays.
I'd like to try NixOs due to it's unique configuration ability, however on a headless server it was a buggy pain just weeks ago (for example user passwords just vanished/changed without any external influence, not allowing access anymore), so i'm open to alternatives.
What i am looking for in advice is:
- a minimal, configurable (file based for git) tiling window manager
- a top status bar like you see in the screenshot that i can freely configure
- as much terminal emulator based as possible (i honestly mostly only need a browser and the terminal, most other apps have a TUI that i can use with the keyboard, see the above requirement)
- terminal based package management as easy as brew (maybe Nix?)
- custom keyboard layout (I am not a native english speaker, so i mapped all non-english characters to my option keys with the english layout as the base)
- Option to use 2 keyboards at once (come by default when using Karabiner Elements) as i combined 2 small keyboards to one to a fake split keyboard ;)
My current stack on macos is Hammerspoon for heavy customization, Karabiner Elements, yabai, kitty (and alacritty, for ssh, as kitty is bad with ssh in my personal experience), sketchybar. firefox (customized for privacy)
Any good recommendations or dotfiles? Anything i should look out for as a MacOs User?
Thanks in advance!
I don't know how the code is currently working, but I like this feature idea and would suggest to start very simple and proceed from there.
For example you could: a) Make a list of communities that are siblings with their id and instance b) add a toggle to view sister community posts yes /no c) query all communities, list the last x posts from each with time constraints, e.g. not older than 1 day or hour depending on the community post frequency d) list them sorted by time of x , depending on what was chosen
The biggest issue I see with this simple approach, besides others, is that different communities are different in terms of activity / post frequency. So ideally the better, but more effort, way would be to let each community instance communicate their posts themselves via a query with activity metric parameters. Basically the amount of returned posts would depend on common parameters set by the most active instance.
It's not yet thought out, but just getting an mvp started and test the waters would probably be better than having it perfect right away while working on it for months
The issue was much more straightforward than i thought. It seems sometimes thinking of too complex issues will hinder finding the easiest cause - the local forewall on the pi was blocking it / had no explcite allow.
To check i did: sudo ufw status verbose
There was only port 22
I added the new port as Allow Port 8081: sudo ufw allow 8081
And it works now! Thanks for all the tipps that pointed me in the right direction!
Thanks for the hints, this definitely helped, however it did not solve the issue.
What i did:
- I changed via
omv-firstaidthe omv port from 80 to 8081. - I confirmed with
ss -ltnthat this change was successful and i see the listening port 80 vanished, while this now popped up:
State Recv-Q Send-Q Local Address:Port Peer Address:Port
LISTEN 0 511 0.0.0.0:8081 0.0.0.0:*
- I tested locally via ssh from the pi the connection via
curl http://mylocalip:8081/and it works, i get the html back - I tested from my laptop (connected to my router via WiFi, where the raspberry is meshed into via the repeater in between) and i still get the timeout.
- I tried tunneling again via ssh
ssh -L 8081:localhost:8081 pi@raspberrypi.localand i did not get any errors this time. However when i open the local url in the browser i get a connection reset and my terminal shows mechannel 3: open failed: administratively prohibited: open failed. However this just says that TcPForwarding is disabled, which is fine, so that tunneling issue should not be the main problem, i assume.
Hello!
I've been running into an oddity and i can not find the root cause.
Situation
I have installed OMV on my raspberry pi 4 4GB via:
wget -O - https://raw.githubusercontent.com/OpenMediaVault-Plugin-Developers/installScript/master/install | sudo bash
I also needed to use usrmerge before the installation:
sudo apt install usrmerge
After completion, while being connected via ssh, i can query the omv website and it works fine:
curl localhost
However whenever i try to access it via the browser, it does not. I have ran omv-firstaid as well just to be sure, but that does not change anything.
Network
My Network is connected via ethernet to a repeater (Fritzbox 4040), which in turn connects to the router via ethernet (Frityzbox 7490). Another repeater is also connected.
All are connected as a singular Mesh.
Question
I can connect via port 22 to my pi from anywhere in my house. It works fine and stable due to the mesh. However i can not connect to port 80 for OMV.
I've tried port forwarding on my network mesh, but that did not change anything.
I also tried for testing purposes a tunnel via ssh ssh -L 80:localhost:80 pi@raspberrypi.local but that resulted in a:
bind [::1]:80: Permission denied channel_setup_fwd_listener_tcpip: cannot listen to port: 80 Could not request local forwarding
Which makes me think it might be the network on the pi. However I am new to linux networking and therefor would like to ask for your ideas.
Any ideas on what could be cause?
Thanks in advance for the help!
(Crosspost from lemmy.ml )
Hello!
I've been running into an oddity and i can not find the root cause.
Situation
I have installed OMV on my raspberry pi 4 4GB via:
wget -O - https://raw.githubusercontent.com/OpenMediaVault-Plugin-Developers/installScript/master/install | sudo bash
I also needed to use usrmerge before the installation:
sudo apt install usrmerge
After completion, while being connected via ssh, i can query the omv website and it works fine:
curl localhost
However whenever i try to access it via the browser, it does not. I have ran omv-firstaid as well just to be sure, but that does not change anything.
Network
My Network is connected via ethernet to a repeater (Fritzbox 4040), which in turn connects to the router via ethernet (Frityzbox 7490). Another repeater is also connected.
All are connected as a singular Mesh.
Question
I can connect via port 22 to my pi from anywhere in my house. It works fine and stable due to the mesh. However i can not connect to port 80 for OMV.
I've tried port forwarding on my network mesh, but that did not change anything.
I also tried for testing purposes a tunnel via ssh ssh -L 80:localhost:80 pi@raspberrypi.local but that resulted in a:
bind [::1]:80: Permission denied channel_setup_fwd_listener_tcpip: cannot listen to port: 80 Could not request local forwarding
Which makes me think it might be the network on the pi. However I am new to linux networking and therefor would like to ask for your ideas.
Any ideas on what could be cause?
Thanks in advance for the help!
The jbod idea sounds good to explore further, as it tha home server and storage would be separated. However it would add an additional device to the power bill.
However i don't need the full amount of all disks at all times. If i'd want to unplug via shell script, i'd need to plug it manually in person back in for storing things. I actually do not need it running all the time, as the home server ssd can cache most of what i need recently in access. The jbod is then more an archive.
i'm mainly looking for a way to power down the inexpensive hdd's. I could use the raspberry pi as the jbod controller, but it does not properly support wake on lan, so thats also not an option
Hello!
Recently i've been planning a home server with proper storage ability, but I got stuck at the details. I hope the experience in this community can help creating a proper plan on how to create my own (first) home server.
Motivation I've been using Cloud Services, especially Google Photos, for years now in order to store and share what i needed. However recently i'm trying to remove my dependency on cloud services and companies significantly, also with data privacy in mind. Therefor i'm planning to self host what i need.
Requirements
Energy Consumption I'm from a country where the energy prices exploded compared to the average salary. So the enegery consumption of the whole system, including storage and others, should be very low. i'm thinking 10 Watt or less in idle. I know it is challenging, therefor i'm asking for support.
Storage I actually do not need much. I estimated that 2TB are sufficient at a minumum. It should be redundant at least 1x, maybe with an additional cold storage added to the mix.
Compute Once i have a home server, it will not end with just storage. i'll probably have to run services like NextCloud or maybe some in direction of Home automation. I also might want to add security cameras to the mix (not long stored, just in case of incidents)
Initial Thoughts and Ideas
My initial idea was to use the Intel N300 with 8 cores and 6W TDP, however that chip does not seem to be released anytime soon. Therefor I was looking at the N100 with 4 cores and the same TDP. The N305 from Intel is interesting, but has higher energy consumption.
Next i looked at the mainboard options. There are some interesting MIni PCs with the N100, including some passive options, but they would only enable me to have max one M.2 and one SATA 2,5" drive, so I would need to see how i can add redundancy to it (or simply use both of these as SSDs with sufficient storage and call it a day, while adding a slow usb drive for additonal storage) There are also new mainboards from Asrock with N100s coming out, which look interesting as well, but would add costs for all the hardware required to run it.
Price wise HDDs would be the better option, however they since they use more energy, they might not be the ideal way for this build. I thought about it and if i add an SSD as the main storage which acts as a cache, i can use the hdds as simple backups. As i do not need them most of the time (once a day backup update would be suffficient) I could in theory power them off most of the time and only power them on for backup updates when needed. Do you know of any ways to handle this automatically for example via shell script? Most of what i found would require me to manually plug them back in via USB.
As for operating system etc. I've been thinking about Proxmox, TrueNas, NixOS and others. as i do not need VM's, most of my services will be available a container, the TrueNas with Docker / Podman Containers might be ideal. Adding full VMs to the system with only an N100 seems insufficient.
RAM: Most systems with the N100 go up to 16 GB. However I think 32GB would be better with many containers just to be future proof. It might be overkill.
Wake up on LAN support might be interesting too. I'm not sure if that would be supported by my choices so far.
Usage
Initially, the system would not be used most of the time. The most often action would most likely be syncing photos from iOS or Android automatically to storage (any recommendations for that?) as a picture or 2 migh be added daily. Everything else is infrequent, which the highest traffic on the system due to looking at the pictures.
Once it's running, this might change of course.
Summary
In summary, i'm just starting to plan this server and I would appreciate it, if you could give me some guidance on how to implement it the best way or point out mistakes i did in my thought process so far.
I also have a Raspberry Pi 4 that i can add to the mix, however it only has 4GB RAM, so it won't be able to do much alone.
