This kind of end to end documentation is invaluable when you're trying to troubleshoot.
I had a service complaining about the root file permissions which changed between Fedora versions [1]. I was never able to figure out what produced the file system image but thanks so this guide I know where to start looking now.
So as a relatively fresh Fedora user (2024) who just sold their Windows PC (2025) and is in love with the OS, where would be a good place to look for areas to contribute in? Is there a place where Fedora maintainers write down what parts of the pipeline are in need of volunteers?
Personally I feel a good way to get involved is to test updates in bodhi for those packages you care about, or to take part in the process that happens around release time where we go through several freeze cycles and have 'blocker bugs' that need their fixes to be validated: https://fedoraproject.org/wiki/QA:SOP_blocker_bug_process
While probably true, I'm not sure how much Fedora can be owned really. We should give credit where credit is due though: IBM has, together with Intel, Red Hat, and a few others, been very supportive of the Linux ecosystem since early on. Their scalability work, their GPLing of code, and upstreaming of said code, did a lot for people to look at Linux seriously. There are obviously people inside IBM who understand this stuff. So there should be hope.
Techrights are hard to take seriously, the site's author has such a bad case of that polemic tendency that he lost a libel case because he wouldn't stop harassing someone (https://caselaw.nationalarchives.gov.uk/ewhc/kb/2025/3063).
> Every build starts from a clean room. You can never get a different result because someone installed something on the builder last week.
At least 6 years ago, this was not entirely true: a package happened to build only because, by happy accident, the builder had a dependency installed by something else. Since the dependency was not in its Build-Requires:, the build would fail in COPR, which really used a clean room. A case of being saved by the alphabetic order, I guess?
Mmm, COPR and Koji I think use the same mock configs but someone else would have to confirm that :)
It's definitely possible that one of a packages build dependencies pulls in something that a package /should/ depend on and then when it drops it suddenly the build fails. That's why I think we mostly want 'autogenerated' BuildRequires nowadays that are generated based on the upstream source or packaging format to make that less of a figuring out excercise.
The best example of suck, and that forced you to restart Linux machines after an update, something that was not the case before... you little ducklings...
Imho its not necessary for the average user, adding too much complexity.
For example, I tried to config gpsd to use another port, nothing would work; systemd quietly overrides gpsd's configfile and required additional systemd incantations to change what should have been easy. And the man pages are a mess
And journalctl makes me want to pull my hair out; just let me grep a regular file - discovering namespaces is a nightmare
Personally I think moving away from the hodgepodge of scripts, giving a consistent interface for people to use, extend, etc, is kinda necessary for the average user to have a decent system. It can often be complex to make things simple.
I'm also not sure how systemd could override a config file, because systemd does not write the unit files on a system. Wouldn't that be on the maintainer, distro, or other methods of getting the unit file?
It sets up the network for services, but it's been a minute since so I don't remember the specifics, just that I needed to make a new file in some deep folder to tell it to use a non-standard port.
To me its just more complexity, and not super consistent, which is fine for enterprise, or automation. I'm used to openbsd, regular shell init scripts go in /etc/rc.d, and my logs are in /var/log. Can't get much simpler than that, but i understand its limitations, and Linux is certainly more of a hodgepodge
It’s at odds with the philosophy of do one thing and do it well. Its monolithic feature set pushes out smaller contributors and leads to increased reliance on a singular package. This tends to self reinforce. I also use systemd in my current distro so not sure where I stand on this but it’s a fair concern.
Before systemd I had Redhat machines with 3 to 4 years of uptime, and now you need to restart Linux after updates....while Windows people rarely and they laugh at you...
If you use the command line to do the update there is no need to reboot, however, the issues that the offline update is trying to solve still stands.
Some programs can crash when all their files are suddenly swapped for new ones. I've seen multiple linux systems bork themselves during an update because the update itself caused the desktop session to crash, which caused the update process to be interrupted just as it was modifying some crucial files, thus rendering the system unbootable.
The systemd init isn't half an operating system any more than the coreutils "ls" utility is half a userspace. The systemd project has many different components but they're independent, they are not all literally part of the init system.
You're the one who brought up init systems. Like you said, systemd is a huge project; an init system is just one of its components.
> he systemd project has many different components but they're independent
They _are_ dependent. It's one codebas, and even if you somehow separated it - you would still only have systemd-specific, systemd-dependent replacements for _actual_ independent components.
They all depend on systemd and won't work with anything else, nor on any other operating system either.
>> The proof this site is misnamed is that a post criticizing systemd gets downvoted, probably by people who have never questioned why an [[ init system ]] needs to become half the operating system.
So no, I didn't bring it up.
I find the whole argument kind of dumb anyway. People talk about this as if it was a violation of the "Unix Philosophy" when Unix and BSD are developed even moreso in the same way - the init system is sitting in the same repo as the libc, the kernel, and the entire rest of the OS. Linux is alone amongst operating systems in making core system components swappable like to that degree in the first place. It's fine to like that kind of flexibility, but a bit silly to act as though that approach is fundamentally superior. If you want to use non-systemd tools, use a non-systemd distro, that is your choice. Just quit whining about 99.9999% of the other software users on the planet not catering to it.
> If you want to use non-systemd tools, use a non-systemd distro, that is your choice.
No, it is not my choice, because I'm not talking about my PC at home, but rather what organizations use - commercial companies, universities, schools, NGOs etc. And also what users are taught to handle and work with.
But it's not even that: I want systemd to be _dropped_ as a default part of the main Linux distributions. I want only the people who explicitly prefer systemd, to use systemd, and not the rest.
As for the "dumb argument" etc. - what you're saying is that because the different BSD distros are monolithic, then not only should Linux become that way, but - all of the lower-level Linux userspace, above the kernel, should just be handed over to the benevolent oligarchy of the systemd people.
That's a bad idea, both in principle, and also in practice, due to the innumerable problematic design and implementation choices systemd has taken - which developers and sysadmins/power-users did not mandate, agree to or support.
> No, it is not my choice, because I'm not talking about my PC at home, but rather what organizations use - commercial companies, universities, schools, NGOs etc
How much of a say do you have over the minutiae choices of various organizations anyways?
> I want only the people who explicitly prefer systemd, to use systemd, and not the rest
And some those people that prefer systemd happen to have a lot of influence on the choices for what init system a distro uses.
Maybe we could build something that lets us convert between different init systems to ease the maintenance burden. We'd need a structured format to start from, I wonder what options are out there...
I think the systemd people state that it is more than an init system, maybe your expectation of it needs to be adjusted. "systemd is a software suite for system and service management on Linux built to unify service configuration and behavior across Linux distributions." One of the great things about Linux is that you can always choose a distro that fits your ideals (or whatever else).
It's absolutely off-topic to moan about systemd on every post about Linux.
If there was a complaint about how systemd complicated builds and tests in a specific way the commenter could name, that'd be valid. Just finger-wagging is tiring after how many, 10 years?
We moan about Microsoft's horrid choices with its operating choices, already for, what, let's say 40 years now? 45? When they cut it out, start using standards, stop forcing crap onto their users (not all of Windows is crap; a lot of it is quite solid), stop with the spying and surveillance, and of course make the code FOSS, then we'll stop the moaning.
It's similar for systemd. It's a poor architectural choice and it needs to be undone. And that's ignoring the personal drama surrounding it over the years which, indeed, needs to be let go by now.
As for topicality: This post is about the "sausage" that is a distribution. Following this metaphor and mentioning a problem with a certain kind of sausage is on-topic.
I had a service complaining about the root file permissions which changed between Fedora versions [1]. I was never able to figure out what produced the file system image but thanks so this guide I know where to start looking now.
[1]: https://bugzilla.redhat.com/show_bug.cgi?id=2402944#c1
Personally I feel a good way to get involved is to test updates in bodhi for those packages you care about, or to take part in the process that happens around release time where we go through several freeze cycles and have 'blocker bugs' that need their fixes to be validated: https://fedoraproject.org/wiki/QA:SOP_blocker_bug_process
https://web.archive.org/web/20230112002240/https://beefymira...
At least 6 years ago, this was not entirely true: a package happened to build only because, by happy accident, the builder had a dependency installed by something else. Since the dependency was not in its Build-Requires:, the build would fail in COPR, which really used a clean room. A case of being saved by the alphabetic order, I guess?
Maybe they have updated Koji since.
It's definitely possible that one of a packages build dependencies pulls in something that a package /should/ depend on and then when it drops it suddenly the build fails. That's why I think we mostly want 'autogenerated' BuildRequires nowadays that are generated based on the upstream source or packaging format to make that less of a figuring out excercise.
https://suckless.org/sucks/systemd/
https://without-systemd.org/wiki/index_php/Arguments_against...
For example, I tried to config gpsd to use another port, nothing would work; systemd quietly overrides gpsd's configfile and required additional systemd incantations to change what should have been easy. And the man pages are a mess
And journalctl makes me want to pull my hair out; just let me grep a regular file - discovering namespaces is a nightmare
I'm also not sure how systemd could override a config file, because systemd does not write the unit files on a system. Wouldn't that be on the maintainer, distro, or other methods of getting the unit file?
To me its just more complexity, and not super consistent, which is fine for enterprise, or automation. I'm used to openbsd, regular shell init scripts go in /etc/rc.d, and my logs are in /var/log. Can't get much simpler than that, but i understand its limitations, and Linux is certainly more of a hodgepodge
ls has more flags than a Soviet military parade.
Some programs can crash when all their files are suddenly swapped for new ones. I've seen multiple linux systems bork themselves during an update because the update itself caused the desktop session to crash, which caused the update process to be interrupted just as it was modifying some crucial files, thus rendering the system unbootable.
> he systemd project has many different components but they're independent
They _are_ dependent. It's one codebas, and even if you somehow separated it - you would still only have systemd-specific, systemd-dependent replacements for _actual_ independent components.
They all depend on systemd and won't work with anything else, nor on any other operating system either.
>> The proof this site is misnamed is that a post criticizing systemd gets downvoted, probably by people who have never questioned why an [[ init system ]] needs to become half the operating system.
So no, I didn't bring it up.
I find the whole argument kind of dumb anyway. People talk about this as if it was a violation of the "Unix Philosophy" when Unix and BSD are developed even moreso in the same way - the init system is sitting in the same repo as the libc, the kernel, and the entire rest of the OS. Linux is alone amongst operating systems in making core system components swappable like to that degree in the first place. It's fine to like that kind of flexibility, but a bit silly to act as though that approach is fundamentally superior. If you want to use non-systemd tools, use a non-systemd distro, that is your choice. Just quit whining about 99.9999% of the other software users on the planet not catering to it.
No, it is not my choice, because I'm not talking about my PC at home, but rather what organizations use - commercial companies, universities, schools, NGOs etc. And also what users are taught to handle and work with.
But it's not even that: I want systemd to be _dropped_ as a default part of the main Linux distributions. I want only the people who explicitly prefer systemd, to use systemd, and not the rest.
As for the "dumb argument" etc. - what you're saying is that because the different BSD distros are monolithic, then not only should Linux become that way, but - all of the lower-level Linux userspace, above the kernel, should just be handed over to the benevolent oligarchy of the systemd people.
That's a bad idea, both in principle, and also in practice, due to the innumerable problematic design and implementation choices systemd has taken - which developers and sysadmins/power-users did not mandate, agree to or support.
How much of a say do you have over the minutiae choices of various organizations anyways?
> I want only the people who explicitly prefer systemd, to use systemd, and not the rest
And some those people that prefer systemd happen to have a lot of influence on the choices for what init system a distro uses.
Maybe we could build something that lets us convert between different init systems to ease the maintenance burden. We'd need a structured format to start from, I wonder what options are out there...
If there was a complaint about how systemd complicated builds and tests in a specific way the commenter could name, that'd be valid. Just finger-wagging is tiring after how many, 10 years?
It's similar for systemd. It's a poor architectural choice and it needs to be undone. And that's ignoring the personal drama surrounding it over the years which, indeed, needs to be let go by now.
As for topicality: This post is about the "sausage" that is a distribution. Following this metaphor and mentioning a problem with a certain kind of sausage is on-topic.