Debian with the docker convenience script.
They seem to be moving away from this, and it’s not longer the first option on their install page
On their debian page
Use a convenience script. Only recommended for testing and development environments
Also, it should be noted about the first option they recommend, Docker Desktop, that Docker Desktop is proprietary.
I recommend just getting the docker.io and docker-compose from debian’s repositories.
Wish I could transcend into declarativity but the thread’s nix survivor ratio is grim
Yeah lol.
I will say, that for my server, I decided to use kubernetes + fluxcd for declaratively. My entire kubernetes “state” is declared in a git repo, and this is the popular, industry standard for things like this, called GitOps. It makes it very easy to add an app, since it’s just adding a folder + some new config files. And unlike Nix, Kubernetes and Flux are very well documented with much tooling as well. Nix doesn’t really have a working LSP or good code autocomplete, but with kubernetes, I can just start typing in a yaml file and then hit tab and it spits out the template for me. Code autocompletion with kubernetes feels much more similar to the tooling of other, more mature tooling
It’s not as declarative as nix though. There are things missing, like OCI containers could theoretically shift if you don’t rely on hashes and some other nitpicks. But declarativity is a spectrum, and I feel like, outside of scientific scenarios (think simulations where versioning, hardware, runtime etc being the same is very important), I think many non-nixos solutions are declarative enough.
Advice online seemed like i needed to basically create a nix flake for the app. I still havent gotten it installed because i have no idea what nix flakes are.
So, the problem is that flakes are technically an “experimental” feature, and thus are not allowed to be included as a primary solution in the official documentation. But, basically everybody uses flakes, so it leads to this crazy documentation split, and is a big part of why documentation on Nix is so part.
Some stuff can only be done with flakes, some stuff only with non-flakes and you have to figure out which is which on your own, while also dealing with the poor documentation for either.
The advice you received was wrong. You could also use a combination of a default.nix file and a shell.nix file to create a package and development environment for your app. But, the documentation is so poor that it’s unlikely you will learn this, and figuring out how to do this on your own, is again, a massive time sink.
So, I use Arch, but I don’t use the AUR at all. Instead, I use nixpkgs to get stuff (admittedly only like 3 packages) not in the Arch repos.
The main reason for this is the quality of AUR packages. Although I don’t really fear a malicious package, I do remember hearing about a package that moved a users /bin to /opt during the install phase.
Something like that is literally impossible with Nix, due to the way that applications aren’t really installed to the system. But, nixpkgs also requires some level of vetting the package quality, which is also nice.
I also use nix for managing all my development environments. For example, my blog github repo, has a few nix files at it’s root, and you should just be able to type nix-shell in folder, and then you will get an identical environment to me.
declarative rollbackable immutability sounds really freakin’ AWESOME
I have BTRFS snapshots set up, and with grub-btrfs, I can even boot from them and revert to an older kernel (my /boot is stored on BTRFS).
However, I have given up on NixOS, for many reasons. The documentation is very poor, and it’s more complexity than it’s worth, to make my whole OS reproducible, rather than just my development environments. In addition to that, their are also issues with running certain apps that expect to see a normal FIlesystem Hierarchy, which nix does not provide. Although you can work around this with stuff like steam-run or creating a fake FHS using nix, I would rather not play that game.
But, considering I installed some stuff in an Ubuntu 22 distrobox recently, because that was what VScode and Unity official provide repos for, maybe this doesn’t really matter. You can probably use distrobox on Nixos, but I’ve seen issues about GPU acceleration with distrobox (and other non-nix apps) as well.
Yes. Firstly, it’s about release cycles. Centos Stream is a rolling release distro (although it rolls very, very slowly). But what this means, is that there isn’t a true guarantee of application/ABI/API compatibility between current versions of Centos Stream and future versions.
In constrast to this, Centos 8 and previous were complete clones of Red Hat Enterprise Linux, which was a stable release distro. During the 10 year lifecycle of each RHEL release, there was a guarantee certain application/ABI/API compatibility not changing, which is what stability in the Linux/software world is defined as.
Centos 8 was a free alternative, for institutions unwilling, or unable to pay for RHEL stable releases. But, with the death of Centos, an alternative was needed. Alma Linux, Rocky Linux, and Scientific Linux (designed for labs and universities), were rebuilds of RHEL. This meant that, they would take RHEL’s open source code, and recompile it and distribute it in a way that guaranteed application/ABI/API compatibility with RHEL, for the same lifecycle of a RHEL release.
So Alma Linux and Rocky Linux fill that gap… but recently, RHEL said that they are adjusting policies to make it much harder for people to make rebuilds (likely targeting Oracle Linux, which is a RHEL rebuild), but this change may affect Alma and Rocky as well.
Rocky said they were going to keep bug-for-bug compatibility, like they used to, but Alma says they are going to do something different. Although they still intend to be ABI compatible, Alma has decided to make some changes to the base system, such as reimplimenting and continuing to support things that Red Hat saw unfit to continue existing in RHEL. One example of this is SPICE, which is a graphics protocol used for low latency display of virtual machines. It had many usecases, and I am very excited to see it back in a distro in the Red Hat ecosystem.
I honestly don’t know how this could turn out.
It could be an amazing change that results in much more progress for hardware acceleration on guests of various types (since that is what vmware is good at) in kvm…
Or it could mean that they are dropping that feature from vmware altogether.
Regardless, I like this change because it means I would be able to run vmware machines and libvirt kvm machines at the same time, at least when I am forced to use vmware workstation.
I also dislike proprietary software in general, so I think less proprietary software and more FOSS is a good thing.
I found this: https://github.com/tenclass/mvisor-win-vgpu-driver
But it is for another foss kvm based hypervisor called mvisor.
I disagree, because they are not the same thing.
Immutable means read only root.
Atomic means that updates are done in a snapshotted manner somehow. It usually means that if an update fails, your system is not in a half working state, but instead will be reverted to the last working state, and that updates are all or nothing.
I create a btrfs snapshot before updates on my Arch Linux system. This is atomic, but not immutable.*
There is also “image based” which distros like ublue (immutable, atomic) are, but Nixos (also immutable and atomic) are not.
*only really before big updates tbh, but I know people do configure snapshits before all updates.
As an alternative suggestion, consider using a linux laptop with a drawing tablet.
I use a wacom intuos s with bluetooth to takes notes on xournal++, although rnote should work as well.
Not infinite ram. I’d say double ram, plus there is a noticable, but quick delay when switching to an application that was compressed by ram. But it’s much, much faster than switching to an app that was swapped to disk.
Cachyos (arch based distro) does this hy default.
Winlator is really just termux + proot + box64 + wine wrapped in a neat UI (+ controller support). You can, and people have set this up manually before winlator came along. You’ll either need termux-x11 or vnc for the GUI.
Mobox is a similar project that does this automatically via a script… but I don’t see a license in their github repo, plus they require the proprietary input bridge for touch controls.
Termux recently got moved off of the play store (kinda), and is now only available on f-droid/github, because Google was further locking down what they allowed on their store.
And in addition to that, they recently added a restriction in later versions of Android: “Child process limit”. Although this limit used to not there, when enabled, it prevents users from truly running arbitrary linux programs, like via termux.
Although the child process limit can still be disabled in developer options, it doesn’t bode well for how flexible base android in the future will be, since many times corpos like Google move stuff into the “secret” options before eventually removing that dial all together.
TLDR: Termux has been, and is a thing… for now.
Also, I want to shout out winlator. It uses a linux proot, similator to termux, and has box64 and wine inside that proot that people can use to play games. I tested with Gungeon, and it even has controller support and performance, which is really impressive.
So I don’t know how much you know about the shell, but the way that the linux command line works is that there are a set of variables, called environment variables, which dictate so me behavior of the shell. For example, $PATH variable, refers to what directories to search through, when you try to execute a program in your shell.
The documentation you linked, wants you to create a custom shell variable, called SCALE_PATH, consisting of a folder path, which contains the compiled binaries/programs of scale you want to run.
This command: export PATH="${SCALE_PATH}/bin:$PATH"
temporarily edits your PATH variable to add that folder with the scale programs you want to run to your path, enabling you to execute them from your shell.
You may be interested in this:
https://github.com/brunodev85/winlator
There is also a fork with increased performance, at the cost of some usability enhancements: https://github.com/MrPhryaNikFrosty/Winlator-Frost
No, I think if you’re using the nextcloud all in one image, then the management image connects to the docker socket and deploys nextcloud using that. The you could be able to update nextcloud via the web ui.
https://github.com/nextcloud/all-in-one?tab=readme-ov-file#how-to-update-the-containers
Because forgejo’s ssh isn’t for a normal ssh service, but rather so that users can access git over ssh.
Now technically, a bastion should work, but it’s not really what people want when they are trying to set up git over ssh. Since git/ssh is a service, rather than an administrative tool, why shouldn’t it be configured within the other tools used for exposes services? (Reverse proxy/caddy).
And in addition to that, people most probably want git/ssh to be available publicly, which a bastion host doesn’t do.
So based on what you’ve said in the comments, I am guessing you are managing all your users with Nixos, in the Nixos config, and want to share these users to other services?
Yeah, I don’t even know sharing Unix users is possible. EDIT: It seems to be based on comments below.
But what I do know is possible, is for Unix/Linux to get it’s users from LDAP. Even sudo is able to read from LDAP, and use LDAP groups to authorize users as being able to sudo.
Setting these up on Nixos is trivial. You can use the users.ldap set of options on Nixos to configure authentication against an external LDAP user. Then, you can configure sudo
After all of that, you could declaratively configure an LDAP server using Nixos, including setting up users. For example, it looks like you can configure users and groups fro the kanidm ldap server
Or you could have a config file for the openldap server
RE: Manage auth at the reverse proxy: If you use Authentik as your LDAP server, it can reverse proxy services and auth users at that step. A common setup I’ve seen is to run another reverse proxy in front of authentik, and then just point that reverse proxy at authentik, and then use authentik to reverse proxy just the services you want behind a login page.
OP is on OpenWRT (a router distro), and Alpine. Those distros don’t come with very much by default, and perl is not a core dependency for any of their default tools. Neither is python.
Based on the way the cosmo project has statically linked builds of python, but not perl, I’m guessing it’s more difficult to create a statically linked perl. This means that it’s more difficult to put perl on a system where it isn’t already there, and that system doesn’t have a package manager*, than python or other options.
*or the the user doesn’t want to use a package manager. OP said they just want to copy a binary around. Can you do that with perl?
Not quite a scripting language, but I highly recommend you check out cosmo for your usecase. Cosmopolitan, and/or Actually Portable Executable (APE for short) is a project to compile a single binary in such a way that is is extremely portable, and that single binary can be copied across multiple operating systems and it will still just run. It supports, windows, linux, mac, and a few BSD’s.
https://cosmo.zip/pub/cosmos/bin/ — this is where you can download precompiled binaries of certain things using cosmo.
From my testing, the APE version of python works great, and is only 34 megabytes, + 12 kilobytes for the ape elf interpreter.
In addition to python, cosmopolitan also has precompiled binaries of:
And a few more, like tclsh, zsh, dash or emacs (53 MB), which I’m pretty sure can be used as an emacs lisp intepreter.
And it should be noted these may require the ape elf interpeter, which is 12 kilobytes, or the ape assimilate program, which is 476 kilobytes.
EDIT: It also looks like there is an APE version of perl, and the full executable is 24 MB.
EDIT again: I found even more APE/cosmo binaries:
Debian already has docker packaged. That’s more convenient.