Devuan, anyone?

[Linux-SWAT]

Or simply pull the plug.
But on a system that you have hardened, this kind of bug is still more than a "simple annoyance".

 

[tigerroast]

The weirdest part was yet to come: Switching to another virtual terminal and back fixed the framebuffer so I didn't worry too much, but when I wanted to shutdown my machine later and the system switched to display the shutdown messages they were preceded by whole forum posts I wrote before and several control sequences as if all keypresses received by X were being printed there. That kinda freaked me out - I couldn't find anything in the logs, though.

It's like the gov't keylogger just flumped itself and threw up all over the computer. Maybe that's why Linux users are on the US terror watchlist lol

But yeah, that's freaky as hell. I couldn't even begin thinking of ways to reproduce that.

 

[Letalis Sonus]

Confirmed: systemd ships NSA keylogger

Jokes aside: Has there ever been a proper code review of systemd? Being PID 1 is serious shit after all.

 

[comradekingu]

Ah, okay. Short of chalking it up to freak coincidence, I couldn't tell you what caused the issue. Sorry, man.

As complexity increases, the chance of error goes up or down?

Yeah, there's some issues, quite valid issues, with systemd brought up in the post. Unfortunately, those criticisms were buried in a swath of the type of vitriol Slashdot shit-posters have become known for. The post is rightfully criticized.

You are adding unfortunate levels of pointing out what you mean, specifically.

Hasn't been an issue for Facebook or Red Hat thus far.

No, I'm sure they don't care about uptime or reputation at all. Its not like that is what they actually respectively base their businesses on.

Even for regular people, it hasn't been a problem in the year the bug existed to the point where people the world over were hitting up Debian's BTS or even systemd's Github over system hangs regarding systemd.

So pure luck and how a bug is handled is now what a bug is. I and everyone else will be asking for your reactive powers once we see it turning out otherwise.

It's been, what, 2 years since I've followed the Great Schism left in systemd's uptake, and no one's proven how systemd's not modular. Matter of fact, this year's systemd.conf was quite interesting in that regard. Would systemd be within this ballpark if it weren't?

It is not modular in the traditional unix sense, which has been a concept there for nearly 50 years. How many people thought it was a bad idea in 50 years? How many of them/how much of it left?
Try replacing your init system with anything else on a systemd-box running any regular desktop. Problem?

This is another systemd trigger point I don't get either. Systemd actually has a pretty clean record regarding security issues, but every time a bug makes a splash, the comparisons to X.org or Windows come and the "dump ShitstemD" signs follow. No one wants to dump the Linux kernel over its security issues (which may have actually helped in the Panama Papers leak, can't cite a source for that claim, it's just something I briefly heard on LAS/Jupiter Broadcasting one day).

A forced migration over to something that leaves no tangible recourse, while promising to do so, expanding indefinitely, and sucking. So the polar opposite of Linux, for Linux-people. I don't get it either. It's like people are upset or something.

 

[kaprikawn]

Hi all

Some "interesting" bits about Systemd here and here.

Cheers, Magic Sam

Some more interesting information here

 

[tigerroast]

As complexity increases, the chance of error goes up or down?

Excellent job jumping the gun, bro; we have no idea if that's a systemd bug that'll keep rearing its head or just freak coincidence. Even if it's the former, then it's not a big deal. If it's repeatable, then it's defeatable.

You are adding unfortunate levels of pointing out what you mean, specifically.

If the meaning of the 2nd sentence I responded to didn't change from the 1st, then repeating myself didn't change anything either.

Besides, we both agreed some valid criticisms of systemd were made. SOME valid criticisms.

No, I'm sure they don't care about uptime or reputation at all. Its not like that is what they actually respectively base their businesses on.

Of course they care, which is the reason they use systemd in plenty of their fleet lol

So pure luck and how a bug is handled is now what a bug is.

Not what I meant and you know it. What I'm saying is, if this bug is apparently so severe and widespread, then why isn't it being represented on Github or freedesktop.org as such? Where's the stream of reports citing many users having their computers crash because of this bit of code? Why haven't the anti-systemd iconoclasts been able to fell our lord and savior, Harry Poettering, and cause the systemd sky to fall in the wake of this horrifying bug?

Because this bug is local-only, so if someone else has local access to your computer without your permission, then that should be an even bigger security concern than the utilization of this bug to crash the system. Not like you can "accidentally" type this code anyway. Still, it'll likely be removed from the next stable systemd release, so no harm, no foul.

I and everyone else will be asking for your reactive powers once we see it turning out otherwise.

Set the timer. Can't wait to see that pie on your face.

It is not modular in the traditional unix sense, which has been a concept there for nearly 50 years. How many people thought it was a bad idea in 50 years? How many of them/how much of it left?

So you concede that systemd is modular, which is true, but you still found a way to make it a problem in your mind. I'm not gonna lie; that's pretty impressive. That's Clinton levels of mental gymnastics.

Seriously, through all the rhetoric, you're the only one I found that said systemd is neither modular or monolithic, and the way you came to such a conclusion was making your own definition based on the UNIX philosophy, which, again, doesn't change systemd's modular nature.

Try replacing your init system with anything else on a systemd-box running any regular desktop. Problem?

Not everything's gonna be compatible with everything else, but you think that's the way it's supposed to be. Apparently, systemd-nspawn needs to work on launchd and Upstart in order to be truly modular.

A forced migration over to something that leaves no tangible recourse, while promising to do so, expanding indefinitely, and sucking. So the polar opposite of Linux, for Linux-people. I don't get it either. It's like people are upset or something.

Linux has always been about branching out and exploring the possibilities, passing up possible solutions for the chance at finding better ones. Linux has become entrenched in a good-yet-polarizing solution, which is going to make people upset. If systemd does turn into X.org (betting my money it won't), then the Linux community at large will find/invent a better solution. Hasn't happened yet though. SysV's a corner case solution now. ConsoleKit's dead. uselessd inevitably followed the path it was taking into its own grave. OpenRC isn't being run outside of a Gentoo/Slackware/PCLOS/Puppy/some Manjaro builds and embedded distros (there's more, but they're insignificant).
[doublepost=1475769115,1475768819][/doublepost]

Jokes aside: Has there ever been a proper code review of systemd? Being PID 1 is serious shit after all.

Linus doesn't have any major issues with systemd, and some Linux devs, while wary of systemd, concede that nightmare scenarios with systemd haven't happened.

But yeah, I agree. A code review of systemd and its parts is important. A quick Google search didn't turn up much, but you'd think they already went through that process before distros everywhere began adopting it. Idk

 

[Linux-SWAT]

You still don't understand what a local user is about, and why this bug is critical.
Slackware doesn't run OpenRC.
There's plenty of critical Linux system that don't and will never use systemd. As a real world example, in my last job, the OS in charge of the cloud and bare metal infrastructure deployment.

 

[Penguin-Guru]

Seems like this thread has fallen back into the SysD debates. Please consider moving to another thread called "SystemD Debates", which I have just created (Link-1 below), so that Devuan users can sort through the information relevant to their configuration more easily.

That said, I think it's safe to say none of us are going to change our opinions regardless of what evidence is brought to bear. There is plenty of information out there for those interested, trying to change the opinions of those uninterested is simply a waste of time. Devuan exists to offer an alternative so I hope that people will come to accept differences of opinion. I'll hang a lantern in advance for what is probably a scientifically controversial statement but I believe that people become emotional on this topic because we are afraid that the alternative technology will usurp our preference, thereby endangering our personal comfort. Personally, I find SysD disgusting primarily on principle, but I will not try to convince people to move away from it; a moral decision based on the same principle-- freedom of choice. In fact, I might even recommend it for some use-cases based on organisational circumstances. It is a matter of personal and professional opinion, but the beauty of free software is that we never run the risk of precluding preferences. Where something is not satisfactory, we may change it to suit our ideals. Naturally, less popular software will suffer from less development effort, but the choice is always our to make. In the case of SysD, there are already satisfactory alternatives and a passionate community to support them long-term. Despite the mass-adoption of SysD, Devuan shines pure as an example of our culture's resilience in solidarity. Although I'm not actually affiliated with them in any way, this is why I wanted to make sure it would be available to Pyra users, and this is why I would like to keep the SysD debates and Devuan support threads separate. Devuan is not about hating SystemD.

Link-1: https://pyra-handheld.com/boards/threads/systemd-debates.78306/

 

[bzar]

@Penguin-Guru: Well said.

For my part I only posted here again because of blatant name calling like "shitstemd", an attempt to hijack the term "linux" to only refer to distributions or setups that have or don't have some components in addition to the kernel, and some factual inaccuracies. To each their own, do what you will and let others do the same. It's not nice to hear something you like being slung mud at just because the mudslinger likes something else. If there's new issues with something, informing others about them in a civilized manner should be enough. Calling people or projects names is... neither persuasive nor called for.

There have been some good arguments made on both sides, and we can each choose to come to our own conclusions about them. I think it's only fair to let others live with their decisions as they let you. I have no problem whatsoever there being other init systems and other people using them.

It might be more constructive for those wanting to argue the superiority of their choice to maybe resort to listing the upsides of their preferred choice instead of trying to come up with downsides in others? It is of course important to look at the downsides as well, but dwelling on only those just makes everything look bad. As an exercise, those wanting to argue superiority, come up with sentences describing upsides that neither mention any alternative implementation nor have a negated word, like "unlike" or "no" or "not". No need to post the results here unless you really want to, but just doing this by yourself can make it easier to see the "other side" as people who just like different things than you, not your enemies or lesser people.

 

[ible]

i think these were the original lyrics:

imagine there's no apple
it's easy if you try
no microsoft below us
using software to spy
imagine all the people
living free today

imagine there's no people
saying their init system is best
nothing to kill or die for
and no android to oppress
imagine all the people
living open source

You may say i'm a Pyrate
but i'm not the only one
i hope someday you'll join us
and the world will create as one.

imagine there's a keyboard
i wonder if you can
whose layout has been forged
by a brotherhood of man
imagine all the Pyras
sharing all the world...

You may say i'm a Pyrate
but i'm not the only one
i hope someday you'll join us
and the world will create as one.

old stuff:

Spoiler

imagine no groups of people...
saying their init system is best...

Imagine there's no keyboard
whose layout hasn't been fixed...

 

[Linux-SWAT]

According to this post:
http://linuxgizmos.com/debian-variant-offers-safe-homeland-for-systemd-haters/
Devuan has ARM images, which is a very good news !

 

[comradekingu]

Excellent job jumping the gun, bro; we have no idea if that's a systemd bug that'll keep rearing its head or just freak coincidence. Even if it's the former, then it's not a big deal. If it's repeatable, then it's defeatable.

Nor did I say it was, however if it was simple, that would be easier to troubleshoot. You err on the side of err. Coincidence sounds a lot like entropy doesn't it.

Not what I meant and you know it. What I'm saying is, if this bug is apparently so severe and widespread, then why isn't it being represented on Github or freedesktop.org as such? Where's the stream of reports citing many users having their computers crash because of this bit of code? Why haven't the anti-systemd iconoclasts been able to fell our lord and savior, Harry Poettering, and cause the systemd sky to fall in the wake of this horrifying bug?

You seem to take security lightly, or you don't understand what a local user is. It doesn't reflect well on your claim that people wanting security use SystemD to that end.

Because this bug is local-only, so if someone else has local access to your computer without your permission, then that should be an even bigger security concern than the utilization of this bug to crash the system. Not like you can "accidentally" type this code anyway. Still, it'll likely be removed from the next stable systemd release, so no harm, no foul.

Local user access is not on the same level as what you can escalate to. Sure problems in release isn't a problem for you, because future. However many people operate in the present.

Set the timer. Can't wait to see that pie on your face.

Even in the past, and invariably relating to complexity in the future, you will find yourself not gloating quite as much.

So you concede that systemd is modular, which is true, but you still found a way to make it a problem in your mind. I'm not gonna lie; that's pretty impressive. That's Clinton levels of mental gymnastics.

You don't understand UNIX philosophy, nor did you answer my question.

Seriously, through all the rhetoric, you're the only one I found that said systemd is neither modular or monolithic, and the way you came to such a conclusion was making your own definition based on the UNIX philosophy, which, again, doesn't change systemd's modular nature.

I don't care about your insular argument, because it doesn't reflect my point about UNIX philosophy, which I did not invent.
Contest that if you want.

Not everything's gonna be compatible with everything else, but you think that's the way it's supposed to be. Apparently, systemd-nspawn needs to work on launchd and Upstart in order to be truly modular.

It would make it modular on a system-wide scale. However you cherry picked two dead projects to make yourself feel good.

Linux has always been about branching out and exploring the possibilities, passing up possible solutions for the chance at finding better ones.

If you mean best of breed, I don't see how it passes up much of anything useful.

Linux has become entrenched in a good-yet-polarizing solution, which is going to make people upset. If systemd does turn into X.org (betting my money it won't), then the Linux community at large will find/invent a better solution. Hasn't happened yet though. SysV's a corner case solution now. ConsoleKit's dead. uselessd inevitably followed the path it was taking into its own grave. OpenRC isn't being run outside of a Gentoo/Slackware/PCLOS/Puppy/some Manjaro builds and embedded distros (there's more, but they're insignificant).
[doublepost=1475769115,1475768819][/doublepost]

So you don't care about legacy. Nor do you care about modularity.

Linus doesn't have any major issues with systemd, and some Linux devs, while wary of systemd, concede that nightmare scenarios with systemd haven't happened.

Which would be what? Short of the things that have happened, I wonder…

But yeah, I agree. A code review of systemd and its parts is important. A quick Google search didn't turn up much, but you'd think they already went through that process before distros everywhere began adopting it. Idk

You don't know the history, nor do you care about content.

Edit: You can use what you are served all you want, marginalizing other options, and the reasons for them, is a bad look.