HayadSont

joined 3 weeks ago
[–] HayadSont@discuss.online 1 points 1 week ago (2 children)

I'm pretty clueless at that point. Never seen something like it. Apologies for not actually being able to help out. Hopefully it will be resolved soon, though.

This is probably not the advice you'd like to hear, but I wonder if rebasing to uBlue's Kinoite would make any difference. Regardless, wish you the best!

[–] HayadSont@discuss.online 1 points 1 week ago* (last edited 1 week ago) (4 children)

No Nvidia, just intel integrated graphics. Its a dell ispiron 7500.

Yeah, that shouldn't cause any pecularities.

LayeredPackages: alacritty bat btop distrobox fish flatpak-builder kpeoplevcard light lsd mako neovim parallel python3-neovim shellcheck swaybg swayfx swayidle swaylock syncthing tailscale tmux virt-manager waybar wlogout wlsunset wofi

Hmm..., nothing too outrageous, I think; still, consider rpm-ostree reset to at least rule out the possibility that any of the layered packages are to be blamed. If you're afraid of losing your working deployment, ensure to evoke the sudo ostree admin pin 1 command to pin it. (ASSUMING THAT THERE ARE ONLY TWO DEPLOYMENTS WHEN YOU DO THIS). After you've done the pinning, ensure that you pinned the correct one by checking this with rpm-ostree status; the deployment you wish to pin should state the fact that it's pinned.

journalctl shows nothing. That’s what I meant by no logs, I should have been clearer. It’s not even getting that far. It’s like it’s stuck in grub, but only when I try the new kernel.

Perhaps I had to be more clear and explicit, boot twice:

  • first, into the now broken deployment, wait until the time your system should have booted otherwise. Then, hard reset.
  • secondly, to your still working deployment, then evoke the journalctl -b -1 command. This should, unless something is wrong with how systemd is setup on your system, show you the logs of the previous boot. You could even compare it with your current boot (which can be accessed with journal -b) and see where it diverges, though looking at the logs of the corrupt boot sequence should suffice.

The only deviation from the vanilla kargs have been for troubleshooting this and I did it in the grub editor so they aren’t persistent. I tried removing rhgb changed quiet to verbose and debug and loglevel=7 just to see if anything would happen. It still just hangs

Aight. Thank you for putting in the work so we can rule that out!

[–] HayadSont@discuss.online 1 points 1 week ago (6 children)

Anybody else experienced something similar?

FWIW, I just booted my laptop that runs a Fedora Atomic derivative -secureblue to be precise*- and updated to the 6.14.5-300 kernel. Afterwards, I rebooted and found a still functional system. Just so you know; GNOME is installed on my system instead of KDE Plasma, but I don't think it would have mattered.

Regardless, your issue remains and is rather peculiar. Uhmm..., I'm just thinking out loud at this point, but have you considered the following:

  • Is this a system with Nvidia? If so, could it be somehow related?
  • How 'pristine' is your system? Have you layered a bunch of packages? If so, do you think the issue would persist after a rpm-ostree reset?
  • Beyond layering, I have experienced that kernel arguments could cause issues down the line. So, have you tinkered with any of that? If so, do you think the issue would persist if you would remove all user-added kernel arguments?

Aside from the above, I don't really know what would cause an issue as such.

Or know of a way I can get some more info out of the system

Consider accessing systemd's journaling with journalctl. This should be (easily) accessible on a successfully booted system. In your case, that would be your working deployment.

[–] HayadSont@discuss.online 3 points 1 week ago (1 children)

Wow, Niri has been shaping up lovely since the last time I checked on it.

That reminds me; I should probably revisit Wayfire (and the other Wayland-WMs) as well 😅. Thank you for the reminder!

[–] HayadSont@discuss.online 2 points 1 week ago* (last edited 1 week ago)

To clarify, this is just my list that I shared in the hope that others would follow suit 😅. (Which hasn't been successful so far...) Consider posting yours 😉!

[–] HayadSont@discuss.online 12 points 1 week ago (4 children)

As someone with a perpetual desire for clean system management—even back in my M$ days^[Which I 'dealt' with by factory resetting every few months 😅]—I deeply resonate with the desire to declare the desired state within a config file and treating it as the single source of truth; this is exactly why NixOS with the Impermanence module has captivated me ever since it appeared on my path, like a long-sought truth.

I've only abstained this long due to lacking a spare device for a proper test run that might lead to permanent adoption. Perhaps this summer will finally be the time to take the plunge.

Looking forward to bringing order to chaos at last.

[–] HayadSont@discuss.online 2 points 1 week ago

The expectation that package maintainers should backport security patches rather than simply updating to the latest upstream version is a rather peculiar quirk.

Can't agree more.

[–] HayadSont@discuss.online 3 points 1 week ago* (last edited 1 week ago) (2 children)

Off-topic, but AFAIK the team responsible for PrivacyTools.io's content has moved to PrivacyGuides.org.

PrivacyTools is now (mostly) solely run by its original owner and its content has ceased to be reliable ever since the likes of NordVPN and Surfshark have appeared at the top of their VPN recommendations.

While I wouldn't argue that PrivacyGuides is perfect, it's undoubtedly better than the alternative. And thus unsurprisingly the actual (spiritual) successor of (what used to be) PrivacyTools.

FWIW, PrivacyGuides doesn't recommend Ubuntu.

P.S. while it doesn't tackle as many topics as PrivacyGuides does, privsec.dev offers comprehensive guides on the topics it does. FWIW, they also used to be in the PrivacyGuides team (and perhaps even in PrivacyTools).

[–] HayadSont@discuss.online 2 points 1 week ago

woah, that’s a lot of information.

Yeah 😅, lol 😂. The first draft was even longer; so much so that it exceeded the character limit :P . FWIW, I had written two drafts on two different days until I landed on (another day) with the third and final version.

Atomic distros do sound pretty good.

Yup, I can vouch for that.

I might try one someday,

Please do :P

but I’ve already setup Fedora Workstation and am very happy with it :D

Glad to hear that, fam! Enjoy!

[–] HayadSont@discuss.online 2 points 2 weeks ago* (last edited 2 weeks ago) (2 children)

Thanks for your detailed reply! Note that in the following writing I've chosen to address matters in generalities and oversimplifications for the sake of readability and brevity. No need to drown you in technicalities 😅.

On KDE Plasma vs GNOME, I would like to try both out and see which I like better long-term. KDE Plasma seems a bit more familiar (closer to Windows 10) whereas GNOME is a bit more different but I'm open to using either.

That makes perfect sense. As noted by others, while installing both DEs on the same system is technically possible, it often leads to conflicts and inconsistencies. For the cleanest experience, consider dual-booting with separate installations for each DE. Since you've already installed Fedora Workstation, you might want to try KDE on a separate installation.

a laptop with an Intel i7-1360P. It's one of those 2-in-1 convertible 360 degree hinge laptops.

Your hardware should work well with either DE. For future reference, check the Linux Hardware Database and ArchWiki's laptop entries for compatibility information. Since you've confirmed the touchscreen and rotation are working, you're already past the biggest hardware hurdles!

I would say I'm open to learning how to work with the terminal and customising the distro a bit, but I don't want to do anything too out of my scope. I don't want to spend too many hours setting it up, I'd rather have something that works mostly out of the box :D

Both Fedora and openSUSE nail this balance perfectly. They provide sensible defaults that work immediately while giving you room to tinker when you're ready.

Stable as I don't want to break my system after an update. I still want an up-to-date distro though. I am open to rolling release distros, but to my knowledge those are usually less stable with more breaking changes than fixed release options.

FWIW, this is where atomic distros really shine: they offer both stability and current software through their unique update model.

Also, how are the "immutable" distros from UB different from the "mutable" distros?

First, a clarification on terminology: The term "immutable" is actually a misnomer that Fedora has moved away from. These systems aren't truly immutable – they can change, but in a controlled, atomic way. That's why Fedora now refers to them as "Atomic" systems, which better describes their update mechanism and system architecture.

Fedora offers traditional Fedora and Fedora Atomic for desktops, with Atomic including variants like Silverblue (GNOME) and Kinoite (KDE). For servers with atomic updates, Fedora offers Fedora CoreOS. Universal Blue (uBlue) enhances these atomic systems with additional features and optimizations.

The key differences between traditional and Atomic systems include:

  • A read-only base system where core system files are protected from casual changes, ensuring system integrity. On traditional Fedora, you can easily remove or modify system files with commands like sudo rm -rf /usr/bin/important-file, but this fails on Atomic systems with an "Error: Read-only file system" message.
  • Many system changes are tracked, which enables you to 'undo' these changes and return to a cleaner state. This tracking mechanism is what makes factory reset features possible, which are being investigated for these systems. This is surprisingly rare in Linux, with few exceptions like Pop!_OS and TUXEDO OS.
  • Atomic updates mean changes either occur or don't – no half-updated broken states. On traditional systems, interruptions can leave you with partially updated components (like an updated kernel but broken system libraries), while on Atomic systems, the previous deployment remains fully intact and bootable if an update fails.
  • Deployment tracking keeps multiple system versions you can switch between using commands like rpm-ostree status to view deployments and rpm-ostree rollback to return to a previous state.
  • Rebasing capability lets you change your entire system base without reinstalling. For example, you can switch from Silverblue to Kinoite with rpm-ostree rebase fedora:fedora/42/x86_64/kinoite.

uBlue enhances these Atomic systems with proactive maintenance. A great example: when the kernel 6.13 update introduced regressions affecting flatpaks, uBlue maintainers pinned the kernel to 6.12, protecting all users while traditional distro users experienced crashes documented in Fedora discussions, Reddit, and EndeavourOS forums.

uBlue also offers streamlined setup with better onboarding than almost any other distro, simplified handling of troublesome hardware, and purpose-built variants optimized for gaming, development, and other specific use cases.

Does it just mean that you're unable to change system-level settings and such/break anything with a mistyped terminal command?

It's more nuanced than that. While changing /usr contents is restricted, modifying /etc works the same as in traditional systems (though changes are tracked). The real protection comes from the deployment system - even if something breaks, you can easily roll back to a working state.

What are the downsides to using an immutable distro?

There are some trade-offs to consider:

  • Traditional Fedora uses a single approach for software management, while Atomic systems use a combination of methods including Flatpaks, containers, and layering. This creates a more compartmentalized but potentially complex software ecosystem.
  • There are also some system limitations to be aware of. Native messaging in browsers installed as flatpaks require workarounds, kernel module options are limited to uBlue's own akmods repository, and there's no current support for UKI. Granted, I don't expect you to engage with the latter two anytime soon.
  • Additionally, with uBlue, you're trusting their maintainers alongside Fedora's team, though uBlue is mentioned in Fedora's documentation.

For someone wanting stability without sacrificing current software, uBlue variants (Bluefin for GNOME, Aurora for KDE, or Bazzite for both) offer significant advantages, though standard Fedora remains excellent if you prefer a conventional approach. Your successful experience with Fedora suggests you're on the right track!

[–] HayadSont@discuss.online 2 points 2 weeks ago* (last edited 2 weeks ago) (4 children)

Okay then, new question, for a beginner friendly distro, should I go for Fedora, OpenSUSE, or something else?

OP, consider making up your mind regarding which one between GNOME and KDE Plasma you'd like to use (at least for the foreseeable future). Afterwards, consider answering the following so that we may do a better job at helping you:

  • What kind of hardware are we dealing with? Can we have the specs?

Note that both Fedora and openSUSE may be considered beginner-friendly. Though, there does exist some considerable difference in design ethos between these and say something like Linux Mint; the former two give you a relatively bare system and assume (at least some) responsibility from its user while setting up the system. By contrast, Linux Mint offers considerable more hand-holding. This may of may not be to your liking.

Note, however, that Fedora and openSUSE are far from the worst offenders in this regard; within the spectrum, they definitely belong to the better half as we've even got distros that assume their users are willing to learn an otherwise useless programming language from scratch. (FYI: I love NixOS and I wouldn't want it anyway else.)

Therefore, allow me to ask another question:

  • How much hand-holding would you deem desirable?

There's also the fleet of distros by Universal Blue that some swear by. These operate with a different paradigm; most of its users would describe them as a better alternative for newbies (under certain circumstances). But I digress...

Finally, I have noted how you've pronounced your preference for a stable system. I do think I understand what you mean by stable, but just to be sure:

  • Stable, as in little to no updates except for those related to security? OR Stable, as in not being afraid to bork your system after an update or otherwise (i.e. kinda synonymous to reliable)? OR... Another use/definition that I've missed?
[–] HayadSont@discuss.online 1 points 2 weeks ago (1 children)

The following has been prepared with help from an LLM. The content is basically mine; it only helped me with wording/phrasing etc. Sometimes, my RSI-like pains come up and I can't be bothered to do otherwise. Thank you for your understanding:


I saw wireguard tools, isn't that a kernel module?

The WireGuard implementation has two parts - the kernel module (built into the Linux kernel) and the userspace tools package. This sysext only provides the userspace tools (wg and wg-quick commands), not the kernel module itself.

Although this looks interesting, I have trouble understanding the pro's and cons vs something like flatpak or containers.

Sysexts fill a critical gap in the Fedora Atomic ecosystem that neither Flatpak nor containers adequately address.

While traditional distros let you install packages natively, Fedora Atomic's direct alternative to this (i.e. layering) comes with significant drawbacks - updates take longer, require reboots that disrupt workflow, and can sometimes block future updates entirely. This has been a persistent pain point for users.

Flatpaks technically support CLI tools but rarely package them, and containers are impractical for things like shells (imagine running fish or zsh in a container to use on your host). Similarly, applications like Steam or certain browsers sometimes need deeper system integration than Flatpak provides - which is why projects like Bazzite and SecureBlue install them (read: Steam and Chromium-derivative respectively) natively.

The CLI situation has been particularly frustrating, even for Universal Blue, which has driven much of Fedora Atomic's ever-growing adoption. Their exploration of various solutions (eventually landing on Homebrew) demonstrates how challenging this problem has been.

Sysexts offer an elegant alternative - they provide system-wide integration without breaking immutability or requiring reboots. You intuitively know when to use a sysext versus Flatpak or containers - they're not competing but complementing each other.

They aren't a silver bullet (we'll still need layering for kernel modules, etc.), but for many tools, sysexts provide the solution the immutable OS ecosystem has been waiting for.

view more: ‹ prev next ›