Lennart Poettering just released within systemd the command "machinectl shell" which is used to create privileged sessions isolated from the original session, just like the old Unix command "su". Lennart's explanation was the next:
Well, there were long discussions about this, but the problem is that what 'su' is supposed to do is unclear. On the one hand it is supposed to open a new session and change a number of execution context parameters (uid, gid, env, ...), and on the other hand it is supposed to inherit many concepts from the original session (tty, cgroup, audit , ...). Since that is so loosely defined it is actually a very weird mix of old and new parameters. To keep this a bit manageable we decided to just change the absolute minimum, and that excludes XDG_RUNTIME_DIR, specifically because XDG_RUNTIME_DIR is coupled with the execution of the / audit session and we didn't convert it. Instead, we simply remove it.
In short: "his" is a broken concept. It offers you a kind of shell, and it's fine to use it for that, but it's not a full login, and you shouldn't be mistaken for one.
This has come up many times, but nothing has changed, therefore I close it. I understand that this is confusing and unexpected, but this is UNIX ... »
This video shows the use of the machinectl shell command in Fedora Rawhide (the development branch) with systemd 225 (the future stable release)
Even when the invasive penetration of SystemD into the heart of the Linux kernel? With the complicity of the users, they do not realize that Lennart Poettering wants to replace the entry "su" at the root of the system [root] by his login through his SystemD ?, that way you want to control all users from the root of the system. The big brother ?.
SystemD is actually a second mini kernel and covers the bugs, without providing anything as an alternative to improving the SysV init that is being replaced. If you go into the SystemD source code, you will notice that there are no comments and it is blank, I hope you do not remove this comment.
What is the problem with new developments? If it is not useful later, it will not be used. As simple as that. If sysv was so good everyone would have adopted it. In the end the best / most comfortable is used
You are wrong we are not against new developments, it is against the massive NIH of Lennart and his systemd team, reinventing the wheel and at the same time breaking compatibility with many applications.
A Windows user talking about user control and invasive penetration, what a troll 😀
It is to understand your offensive comment towards those who do not use SystemD, how you use Fedora the Red Hat experiment to launch its new versions, where Lennart Poettering the developer of SystemD works, who is soon to end the system security from the root, introducing his own Control Login, eventually becoming his kernel-LenuX / SystemdOS, to market to the free world, an old Red Had dream.
@worn, the guy who uses Windows says it hahahahaha
That user, uses what comes out of them and perhaps is at that time in a Win out of necessity, apart from what he said is a very great truth, fuck whoever it is.
Source code without comments ... it seems to me that they do not want to facilitate the understanding of the code or encourage participation, because that will be.
And more troll is the one that writes from the most penetrating general use distro of SystemD it has. : v
A couple of little things. First of all "su" is a program that is part of the core of the operating system commonly known as Linux, but "su" is not part of the Linux kernel. Those core programs are GNU.
Second. Replacing is the essence of Linux, not because something has always been going to be considered an untouchable relic, as if it were a cult. If something better comes out, it will be replaced.
Mind you, I'm not saying that "machinectl shell" is better than "su", I just discovered the existence of the first one and I'm quite intrigued. I'm going to investigate, but I still don't have an opinion on which one is better.
In part I agree with you, elsewhere I wrote something similar to what you say. But someone who comes from Fedora looks a bit biased. I come from one of the many "buntus". I'm from Peppermint 6 and what concerns me is the way they are doing it. I am in favor of innovation, but not at the expense of the community. Many people see the fragmentation of GNU / Linux as a weakness, I see it as a strength. If it weren't for fragmentation, who would prevent a great technological empire "you named it" from keeping part of GNU / Linux? As they say around here, it looks hala'u by the hair.
It is a lie like a temple that "someone who comes from Fedora is biased"
And what about me? I use Fedora a few weeks ago, before I was on KaOS, I used Archlinux for a long time, I have a homemade server in which I use Debian 8, and guess what they all have in common? USE systemd ...
And those who already know me know that I use about 1 year Funtoo, (that I migrate from Gentoo) .. and yet I agree with notfrombroklin, the fact that we show some sympathy towards systemd, does that make us "biased"? ….
and .. since they know so much why don't they start forking? (Or are they doing the "Argentinazo"? Attention, I am Argentine, and for those who do not understand what I mean, I will explain: basically in Argentina it is very common for everyone to create "technical directors" that is, everyone "knows what to do "and" how to do it "and" what is the best "but nobody does shit: v hahaha) I started I think Fedora, then Archlinux brought it as a flag, then Opensuse, and when Debian adopted it, everyone lost their minds, they had enough time to "be against" or "express their disgust" but nobody did anything ... so now to put up with it, or following the philosophy of free software, forking, that's what the code is for, I personally have a certain affinity with systemd , but I know the 2 bells, sysadmins that complain about the binary logs of the asshole bugs that Systemd sometimes has (or big like the videos that synflag uploads bursting Systemd), and on the other hand I know sysadmins grateful with the "services" of Systemd that make it easy for you yes, among them the mounting of remote volumes ... what do I know, I'm not a sysadmin, I'm a systems engineering student, and there is the madman I liked the syntax of the units and allowed me to define a dropbox service that under init I could never get it to work
He is concerned about the penetration of systemd to the root of the system, he wants to end the security of the system, and the system depends solely on his "login systemd", without giving details as they did with KDBUS, managing to put in the linux kernel by the employees of Lennart and company, doing a reimplantation of D-BUS inside the kernel, in protected mode, passed the user code to kernel space, but the users remain silent about the problems and busg that it generates in order to defend systemd that is ending with freedom. Soon Linux users will abandon in mace to another system that does not have systemd
it is more than logical to put DBUS or something similar in the kernel .. or where do you think the "responsiveness" of systems like Haiku comes from or what was BeOS? ... http://diegocg.blogspot.com.ar/2014/02/por-que-kdbus.html
I am struck by the position you take in front of something that you do not know Edu .. !! There are more important issues to discuss against it. The linux kernel has so much proprietary firmware that it should simply be your underlying concern if as an intelligent user you appreciate the security of what you do on a Gnu linux machine.
You can't pretend to know a lot if you don't know why things are happening .. !! I suggest first documenting yourself before issuing comments that put your intellectual capacity on trial ...
Or migrate to a non-Parabola FSF sponsored distro.
Good thing FreeBSD exists for those of us who love UNIX. For me, hurting a lot, GNU / linux has died.
GNU / Systemd-Lennux
This is what the future GNU / Linux will be called, the long-awaited dream of Red Hat, for years they wanted to dominate the Linux kernel, little by little they are achieving it towards total dominance for its commercialization.
Hahahahahahahahahahaha this news made my day 😀
Is »your» broken? What kind of stupid is this?
And now the pro-systemd will say it's true, that "su" has always been broken and that the new "su-systemd" is 1000 times better because it has bars and cheap sluts 😀
This is what was really missing to really have too much concern about systemd and what it is doing in GNU / Linux.
Pro-systemd gentlemen who advocate for it due to the idea that GNU / Linux is very fragmented, I'll tell you something, there are better, more communal, less invasive and imposed ways to make these changes, Lennart and company are following a very strange agenda from Red Hat, and this news only makes things even clearer, this is a DANGER. If you want GNU / Linux to have a common base in order to increase interoperability and compatibility between distros, let me tell you that this is not the way, or at least not the best of them. A real shame that this is happening: /
I don't understand the pejorative, confusing, and suggestible devices you use to attack systemd. In many cases they share opinions that include words with deeply negative impressions such as: invasive, inflated, imposing, etc., and even important omissions in the speech.
But the technical reasons are much more profound because the context in which its use, development and administration is considered is really specific. Making use of these terms generalizes easy-to-manipulate aspects to handle individuals emotionally. It is simple when talking about freedom, justice, imposition and even to add adjectives. Looking a bit at your comment and the delimitation you make on the concepts you raise to rate systemd, I could say without fear of being wrong, doing the same evaluation on other projects, that they would also get a similar rating, and even worse. Why? It is not a sober, cold and logical argument, it uses the emotional appeal to convince. The greatest divergences are limited to nonconformities in the design of the software or the project that manages it, because according to certain customs and perceptions it is not the correct method because it is unknown.
This happens here.
Example:
"" His "is broken? What kind of stupid is this? "
Poettering did not say that "his" was broken. But the concept that "his" represents.
Here there was a very important omission of information.
«Gentlemen pro-systemd who advocate for it due to the idea that GNU / Linux is very fragmented, I will tell you something, there are better, more communal, less invasive and imposed ways to make these changes, Lennart and company are following a very rare from Red Hat, and this news only makes things even clearer, this is a DANGER. "
Here systemd is pointed out as an invasive platform, not sufficiently communal, and an inappropriate global enforcement. In the first place, the word "invasive" should be noted, it has a negative connotation because it involves the use of force. But explanations of its architecture have been given many times and are generally justified. The design is clean enough to integrate into the system or change it for the better. To define what is more or less community, we would have to be more specific, but I do not understand why systemd should be treated in such a special way. The membership involved in development and testing is large enough to be considered community. And even if it were not, it is free software. The term "imposition" can often be viewed negatively, but not in other respects. It should be clearer here. It could be an imposition without any kind of bad integration of the Debian installer, by the Debian project. And it is known to work well and meet installation needs for system users and a negative imposition if openSuse decides to add a bad kernel. Many pieces of software can be imposed and there are different ways to do it. In this case everything was accepted by developers with interests very different from those of Red Hat. You have people from Suse, Debian, Arch Linux, Fedora, Mageia, etc. It's ridiculous to create a conspiracy theory for that.
Well, many reasons have been given in other interventions not only mine but also other users, developers and sysadmins in the case of systemd. Talking about it over and over again at this point makes me lazy, because despite the fact that at the time I have also offered compelling reasons, presented bugs without fixing, censorship tests in bugreports, security problems self-generated by the developers themselves, among other things, all that seems to "pass from one ear to the other" to systemd supporters blindly. That my intervention has been very emotional? Well of course I did not intend to make a technical intervention on the subject, talking about technical things here is somewhat complex, for that is the IRC or better the development list.
I clarify that although I think that systemd is an excellent "init" (understand that concept well) with some very interesting features, the systemd developers have taken license and carte blanche to change so many things in GNU / Linux that in the end this " init »(if it can be called that at this moment) has become an imposition on their part (creating strong dependencies for current and future developments), because as they develop the system they are making it increasingly difficult and the development of other options may even be impossible in the future, as is the case with Gentoo's eudev project, whose development has been hampered on various occasions by the development of systemd / udev, and of which Lennart has already offered an omen of what awaits them.
systemd can be a very good choice as init, but remember, systemd now has a built-in http micro-server, ssh capabilities for remote login, dns, mdns, dhcp, nspawn, logs, dbus, mount, inotify, swap handling , management of quota, integration with XDG_RUNTIME, selinux, pam, integrated management of SSL, comprehensive management of seed and snapshots, little by little it is integrating more and more with btrfs, management of / dev / random (if the same that gives us secure keys: D, conspiranoicos run away), integration with policykit, login, kexec management, and now a privilege escalation coupled with all that.
Greetings.
Opinions, in general, are subjective findings (another issue is, for example, public opinion). They have an argumentative character, and can be positive and negative. They pursue the truth, but cannot affirm it. Instead, they presuppose its opposite, the existence of "other" opinion. It is normal for them to use terms that on the contrary seem pejorative and "deeply" negative. Delving into critical debate can cause these kinds of impressions and feelings. This should be easily understood.
For someone to find the rival's arguments "confusing" may be the same strategy of disavowing the contrary. It is a technique like another to discredit him (it seems to me that this is the case at hand, because Yukiteru's opinions are not confusing)
The boundary between suggestion (an elegant way of talking about manipulation) and the intention to convince, can be blurred depending on the color of the glass from which it is viewed. It re-determines for us where a maneuver is used. In this case, Yukiteru has been one of the individuals in the community who has argued the most and best the problems and consequences of using systemd. Always from the rigor. His technical judgments have been deep and contextualized (I invite you to follow the forum threads or the comments on this blog so that you can read them and see for yourself). Crossing out his comments as manipulative is a way of eroding his prestige within the collective.
That 'Poettering didn't say that' his' was broken. But the concept that represents 'his' »? Rhetoric. Tell him what you want, but 'su' is a utility, a command, that applies a concept. Please, let's not get sophisticated.
"The explanations of its architecture have been given many times and are generally justified", is a way to drain the bulge and settle a burning issue when the debate has not yet concluded. Resistance to these "justified" explanations is widespread among the community, and their controversy seems endless. There are good reasons to think that systemd "is" an invasive component, however "negative" it may seem. The often vaunted explanations of the systemd team do not convince or inspire confidence in a broad critical sector, which many want to silence with a derogatory dialectic about "progress" and the "minimization" of well-founded criticism.
The conflict with systemd provokes serious disagreements, questioning those elements that articulate the conceptual basis of GNU-Linux. We live in a conflictive reality due to incompatible interests. Incompatible because you want to make systemd an axiom. The axioms in GNU-Linux exist, of course, but there is no consensus that systemd is or ends up being.
Regards,
What he says about "his", which mixes context parameters, is true, but hence calling it "inherently broken" is going overboard. When it was invented it was a marvel and the fact that it works is that it continues to be used until now. Although it is also true that the best is not always discovered / invented / created the first time, this new invention may or may not be something better.
What is clear that criticizing for criticizing is going to get nowhere.
Out of curiosity, not only you, Yukiteru, but all of you who are so against systemd, you know that systemd is more and more embedded within the startup systems and that it is increasingly difficult to remove it, in 10 years when it is so convoluted that it will be systemd or something that is not Linux, are you going to switch to bsd or windows?
If you look at the comment on the top right you can see that he is using a freebsd
For my part, I hope GNU / Hurd 😉
Corrects the «lasrgas» looks ugly D:
And you tell me 😀
The "had" is much worse.
I just came to say that. My eyes still bleed.
"There were" long discussions.
I already corrected both
From my position as a Linux fan (not a fanboy, because I admit that I still use Windows thanks to AutoCAD, the Adobe suite, CorelDraw and certain Word macros that LibreOffice and WPS make balls to open correctly thanks to the stupidity OOXML and its Pottering-like standards), SystemD, as innovative as it may be, I do not see it as the definitive solution at an industrial level, since SynFlag has shown that this INIT has had a beginning full of bugs and glitches that have compromised the integrity from the OS, so when I started testing Debian Jessie in the testing branch, I finally managed to remove the SystemD from it, but happiness was very short-lived during that testing time. At the end of the day, I ended up with Debian Jessie that comes with SystemD accompanied by SysVinit, so it no longer breaks with the paradigm of the learning curve.
On the other hand, the most accessible options would be FreeBSD and OpenBSD, but it would be recommended if your hardware dates between 2004 and 2008, because BSD focuses more than anything on the server aspect (in effect, FreeBSD and OpenBSD are more KISS than Arch and Slackware, so while the modus operandi of installing some tools may vary, the result would be the same or better).
So far Debian Jessie has done a good job keeping SystemD in place, but I hope Stretch will replace SystemD with UselessD, because unfortunately, if you want to know more than what dmesg gives you, it's not as practical as the ones are. logs.
GNU / SystemD-LennuX It's comming
Wouldn't it be better if this man would create his own OS, and stop changing things?
Perhaps their intentions and their contributions are great, but I feel that it is a bit to invade what one uses. Although GNU / Linux is free to modify, I feel that to a certain extent more and more things of this man are being imposed, I am not saying that it is bad but the situation does not convince me, gradually we are already with SElinux, SystemD now this: /
Maybe I have to drop arch and switch to FreeBSD 🙁
It seems dangerous to me that a component that is supposed to be part of the system, has its own permission system.
This type of movement makes me see something more and more clearly. If things do not change, I am very afraid that after a few years GNU / Linux will be dead and only systemdOS will remain. They are already sticking their tentacles into the kernel with the consent of Mister Torvalds.
If this individual continues like this, he will say that the kernel is also broken and must be changed to systemdkernel and in this way Mr. Torvalds will be unemployed and the Linux universe as we know it will only be a memory.
and shit. for me it is always sudo su, if they screw that up, I will leave linux.
Years go by and the man continues to cause controversy. I thought that with the fight in 2013-2014 debian and the fall of 8 leaders I was satisfied. "Su" is already dangerous, do we need another one ?, since it leaves the terminal open to root and meanwhile the user runs below, until we do not type exit. "Su" I put that it exists for remote access -obviously it is shell , not graphical applications-, so as not to be using the same password (sudo) to climb privileges in an SSH session. Those things the desktop user does not know or have. Ubuntu has it disabled.
And here we go again ...
People who do not understand what it is about should not give their opinion, unless they are trolls of course.
I, as well as the yukitero user and many others, have tired of noticing red hat plans, so it is no longer useful to repeat like parrots… .there is an old saying that prays… .for a good understanding… few words.
While I recently read a note that I am glad ... there is life beyond gentoo and openrc, incredible as it may seem.
http://lamiradadelreplicante.com/2015/08/30/manjaro-fluxbox-0-8-13-1-dos-sistemas-de-inicio-a-elegir/
I recommend to those who come out in favor of systemd ... read a lot gentlemen and the truth will appear very clear before your eyes.
I'm getting ready to leave ArchLinux and look for a distro without SystemD or, failing that, to migrate directly to a BSD derivative.
It is very sad what is happening with GNU / Linux and that many developers of different distros support the whims of Mr. Lennart Poettering instead of opposing it.
Try Debian-based Antix that comes without SystemD.
😀
I was thinking of Devuan as a viable alternative or Manjaro with OpenRC which would also be a good option considering that the Manjaro team is generating a friendly development with support for SystemD and OpenRC.
you can use arch without systemd, but you can also use linux mint the debian version does not plan to install it and apparently clement lefebvre (its creator) does not like it much either.
The truth is I don't know much about issues since I'm not an engineer or a programmer, but couldn't su be used in systemd to break the system in an easier way?
I think there are good reasons to go against systemd. NOT because of how invasive it is, which in itself is, but the fact that it intermingles with the kernel itself and makes it difficult for other alternatives to systemd to be difficult to work with. I know what I'm talking about and I mean it with my experience in archlinux. I am a user of the Openrc init system and I must say that it does not always break, but there are times that there are things that do not work well and I know that it is because of the changes that systemd constantly requires to be made to the kernel to make Linux more compatible with systemd but at the same time less compatible with other init.
So let's think and take the following question as a reflection, isn't a startup system considered harmful and invasive, if it takes attributions beyond those necessary, but also its objective is to make the compatibility of the primary system (in this case the linux kernel) with other boot systems is more incompatible and problematic? Doesn't that seem like it goes against the principles and ideals of Free Software? Do not you think that the base under which something is done is more important (and this I mean the ideals or intentions of the creator) and not the result is good or bad? (Take for example the discovery of nuclear fission and its use, which depends specifically on the ideal that the user uses it. Be it to create an energy system or a bomb of mass destruction.)
In conclusion, it seems to me that it is. The concept of systemd is not bad, in other words it is a great idea. But the intentions of its creator are really refutable and its ideals do not seem at all close to what free software is supposed to, especially the part of the right to choose mainly, passing the right of the community to decide whether or not something is appropriate for that project for the digestive system, specifically for the waste outlet area (not to say what I really thought), noting that in fact this project is at the service of that community and their opinion is really a priority, as much as it may seem or not and should be very well analyzed and tested before being discarded. However, I find in my opinion that this Man does not care at all about the community on which he depends, it seems that he only cares about the success mainly of him and it is obvious that his success depends on his systemd startup system, regardless of the cost long-term and the impact it may have on the community and other projects.
I have read about the valid reasons why many are against systemd and they seem quite reasonable and valid to me, but this man and I know, not only has ignored them, but also does it with a rather derogatory tone and I dare to say proud and full of egotism. And that is the reason why your startup system is riddled with bugs, strange system behaviors and so many other things that have been overlooked by common users who fall like flies for its easy operation but riddled with such traps and mines waiting for someone I stepped on them to explode.
It seems to me that the concerns of our colleague Yukitero are quite correct and I am afraid that even justified and as a quite pragmatic User, I would say that I share them and that is why I use openrc instead, although the compatibility of the system becomes more and more difficult for me and the use of Archlinux, which I am 100% in love with.
PS: I was happy to give this intellectual tone to the comment, to add a little humor to the discussion, you have permission to laugh if my way of expressing my opinion seems too pompous!: P
You're right and I would add that Lennart is the public face, but systemd is clearly Red Hat's thing and I currently suspect that the NSA might have something to do with it. I think they don't like the fact that there is a popular OS that they can't control (I mean the NSA).
Another comment that makes my day.
I agree with what you say, systemd is a great idea, until the precise moment when you want to do things that do not correspond to what it pretends to be, which is precisely an init.
To those who think I am just talking, I am going to say that I have been using systemd since Arch Linux made the migration to it, and at that time, many will remember the impact it had, many systems dead or with limited functionality due to migration (minute of silence for the fallen localhost ... peace to your bits: D).
Now what I don't like in systemd is that incredible NIH that its developers suffer, trying to reinvent the wheel just because they please and they want to, in front of that I ask a simple question:
Couldn't they make those better ones directly in the original source of their and in that way that the whole community, even those detractors of systemd could enjoy the improvement?
It seems to me that this was a better option than redoing all of its and embedding it as part of the expanding universe that systemd carries with it, making the code not only bigger, but also more complex, difficult to hack (get the term correctly), and with a larger memory footprint. Many will say: "There is plenty of memory."
Greetings.
I'm going to say the same thing that I said on my social networks: I shit on Systemd, Lennart Pototo and all that that entails. I don't know which is worse, whether to use Windows or use this shit .. FreeBSD, I see you .. I see you ..
You are right, to me as a user I notice that Debian Jessie does not go as well as Wheezy and does things that I have no idea why but you are wrong about one thing, it is not that it is worse windows or this, we are already on the way in the future and it's called winux 😛
I just hope that if Devuan one day sees the light they publish the free badge they are talking about and the other distros send this to the full since this whole thing is to make your ears roll.
Sorry, but I prefer a thousand times that Debian Stretch replace SystemD with UselessD for compatibility, since Devuan has not shown interest in using this fork that guarantees 100% compatibility with other INITs such as SysVinit.
We still have free systemd distros. I'm reading about the installation of Gentoo to replace Arch, I also see that it also uses a BSD kernel if you want, I love the idea of having an alternative to Linux in the same distro, I wish we had Hurd as well.
On the other hand with LinuxMint I'll wait to see what they decide. If they definitely implement systemd, I will also look for an alternative for the family.
I saw this news as soon as it came out, I did not write anything so as not to have to bitch, why is it broken? Where has that been seen? Instead of repairing what is "broken" they create something alternative ... that depends on (again) systemd ... Does this move by REDHAT seem strange to anyone? that uses Lennart as a vehicle? Seriously? Tell me no ...
Do you know what is screwed up? That when something really happens with all this "up to" RedHat, Lennart and others, then at that time the alternatives that we already had will start to come out again ... and it will be the riot. As they say around here: Nobody remembers San Pedro until it thunders.
Well look at this: https://forums.freebsd.org/threads/i-present-you-the-next-edition-of-freebsd.52956/#post-297728
Launchd is coming to freeBSD. So it won't take long for rc to disappear from freeBSD as well. And I don't know why systemd is very familiar to launchd ...
It is a very banal argument.
Look at the source of su, man su or red hat documentation and there you have it clear what it does.
It is possible that the some distro does things more things less, but that happens with a lot of free software and that is not why they are written añternatovas of 0.
It is a poor justification, more than anything it seeks the approval of basic users who do not use its or do not know it thoroughly.
There is no worse blind man than the one who does not want to see.
You must be an experienced user who must know how systemd works, you must read the code and understand it 100% or the help of systemd to understand how it works, what it does and what it does not do, to say that the partner has trivial arguments.
To say that something is banal you have to have good arguments, etc ..., so that what one tries to imply or give an opinion is credible.
If we are not lost and we do not contribute anything constructive.
You think it is a good argument not to use "his" if he "doesn't know what he does?"
He does not know what he does who does not want to see how it works, I think that just in a world of free software it is one thing that we cannot complain about, not knowing how something works.
The new solution may be magic, I do not discuss it or criticize systemd, I criticize the argument why make a new command, until I see some advantage over "su" I will continue to use it.
At no time did I refer to any comment, only to the article, just in case.
cat / proc / version
lennax version 4.1.6-1-ARCH
Very original rain. 🙂
lsb_release-a
modules are available.
Distributor ID: Red Hate
Description: LostDebian GNU / Linux (lost os)
Release: Lennax
Codename: lennaxOS
Very original rain. 🙂
lsb_release-a
No LSB Modules ARE Available.
Distributor ID: Red Hate GNU / Linux.
Description: Lost Debian GNU / Linux (Red Hate / LennaxOs)
Release: Red Hate it takes over GNU / Linux.
Codename: lennaxOs 0.1.5
He is concerned about the penetration of systemd to the root of the system, he wants to end the security of the system, and the system depends solely on his "login systemd", without giving details as they did with KDBUS, managing to put in the linux kernel by the employees of Lennart and company, doing a reimplantation of D-BUS inside the kernel, in protected mode, passed the user code to kernel space, but the users remain silent about the problems and busg that it generates in order to defend systemd that is ending with freedom. Soon Linux users will abandon in mace to another system that does not have systemd
What you say is very true, but what can be done?
There is no one or a leader who regulates what is happening, we are to what the gentlemen say and impose because they are contributing according to to improve.
I try to support by donating to the Devuan project and hopefully it will soon see the light of day to make a little opposition to what is being forged with those of redhat / NASA / CIA.
He believed that you have to turn to see The BSDS, as long as they do not think of doing something similar.
They return this green, it depends on a kernel, a fork is difficult to end up connecting to LenuX / Sytemd-kernel. The alarm of the users who comment in the various forums, tends to think of pessimism, every time systemd is penetrating the kenel and the root of the system, which is not good, doing functions that do not correspond to it.
Devuan if you want to be free of systemd, you will have to form your own "kernel fork" which I find difficult, and finally does not see the light.
What will become of ArchBSD?
I see that some here use the concept "GNU / Systemd-Lennux", I think that if this continues like this, there will be nothing left of GNU, they already started with their ... what will follow later?
They treated me as a conspiratorial when I talked a LONG ago about systemd and the REDHAT idea of dominating everything by invading all distros with systemd and that this was going for more. Now it's your, tomorrow will be the logger, ah no, that's done ... well it will be the PAM, ah that's also done, it's called logind ... well the network !, ah no, this too, it's called networkd ... well, just wait to see how GNU / Linux is going to be systemd / Linux. The only thing left of GNU is glibc and something else, because even the FS hierarchy has changed it.
Be careful if Lennart doesn't give up on his April Fool's Day joke and complies with what he said about making his own C library and embedding it in systemd.
It can't be 😀
That detail did not know. Your own C? Library. I'm not saying that glibc is bug-free (segfault when using a regex in grep), and that BSD's libc is no better. But from there to a new one, which would not be serious, but a new made in redhat ?. It's already a lot. In a few years he will make his own kernel with games of chance and sluts ..
Well, if a "little joke" that good Lennart gave himself on the systemd list, but since good Lennart likes to keep his word, even if they are very bad jokes, it is terrifying to think that one day he might keep his word, and be careful, it may be when you least expect it.
I also mention the kernel thing in the same note, but in this case he was more interested in Hurd's microkernel design, so be careful and he does not do the same
http://lists.freedesktop.org/archives/systemd-devel/2013-March/010062.html
Personally, I would not leave the su for the machinectl shell, first of all I do not want to write more to do the same, we go that from the way in which the article is explained and from the comments the end is the same even though it is do it in a different way. Personally I find it ugly, innovative that you can put a pid, but still it seems ugly and impractical, note that personally I like systemd and to date I have not had a single error but I do not want to use and I would not like use that "alternative" neither today nor in the future.
There is something that is not very clear to me and it is that I am possibly confusing or seeing it from the wrong point of view. Hypothetically speaking, they remove su and leave machinectl shell by default and as the only alternative would you also have to use machinectl instead of sudo? (in a nutshell, if they would encasquetaría the shell instead of su and sudo).
By the way, I must also say that before I did not support systemd because I did not have cron, since I discovered that systemd uses timers personally I left the dislike and addressed a more neutral feeling, I see that many complain about systemd and everything bad that is. They do nothing but complain and personally I can say that if they don't like it they can go to distributions without systemd, UNIX, windows or whatever, but after all, if the case is not to support systemd, don't use it. Of course I am not referring to those BSDians who comment, I mean the haters who use ubuntu, fedora, arch, debian, etc, those conformists are the ones who despite the complaints and displeasure are still there. It is not difficult to make a change, I did it several times for a long time, from windows to mandriva and when it broke I was a victim of distro hopping until I got to arch and kaos. The moment systemd becomes a problem for me, I get out of slack (if it doesn't implement systemd yet), or some rolling without systemd and ultimately to bsd. The average Linux user is supposed to be smarter than that.
Will there be people who will feel "kicked out" because of systemd, to the point of crossing over to Windows or BSD? It is a very passive position, and it is what Mr. Poettering and his acolytes want. That people do not think about systemd and if possible leave the distro (s). He is described as a troll or hater. (John Sullivan of the FSF is one of the "troll" just for asking). This degree of aggression has been reached.
With commercial distros like RHEL or Fedora it can work (money and boss rule). But in debian users can demand compliance with article 4 of the social contract.
I did not want to write about this because I do not use systemd and I do not understand much about the subject either, I am an end user and I do not like where it is arriving, it is not using systemd or it is not messing everything, it is not so easy to get to say I put on this desktop because even this they need it, some things from the init will say but it doesn't seem logical to me. Before I used debian, I used even the last one to give it a try. I don't like talking for the sake of talking (some mistakes… Buf). I started to install salix (derived from Slackware but with slapt-get and sourcery, a kind of front-end to compile from the slackbuild page).
I am very happy with lilo (a shame about the development ..) So you can say that I tested systemd and I did not like it, I just hope and wish questions systems are not so big questions I have to prove bar freebsd or derivatives because of this.
That comment gives me a little bit more insight into what's going on, thanks jrm. In fact I always saw that there were haters and pro-systemd but I really saw that it was not so bad according to my criteria, having seen criticisms in favor and against, the same elav says that his laptop had a problem when he tried to pay it ( it took about or more than a minute and there were no logs about it), I understand that in certain tasks it is perfect, but in others it is a complete slop. Honestly, I am somewhat puzzled and I just cannot objectively understand what it is, what they want to achieve and through what.
ianpocks in fact I hope the same and I would like to continue with many distros as they are for a long time before having to document myself and make the leap to other distributions or other systems. I understand that you can also change systemd to uselessd or some other init, what I do not know is how much impact it has on the use of the system, how good or bad it is for end users like many of us.
I clarify my sources, Stefano Zacchiroli (debian) himself described John Sullivan (FSF) as a troll for asking "RMS, do you have an opinion about systemd?" "No. I know it's free software, so you can make your own opinion about it. » # lp2015 ″
It is notable who add it to favorites zacchiro / status / 579289388208775168
I do not understand why so much alarm. If you don't like SystemD use another distribution, I use Funtoo and it looks great.
regards
koprotk you don't have software limitations like not being able to use gnome 3.8?
@rain is Gnome 3.14.4… http://i.imgur.com/FBiAxoj.jpg
thank you x11tete11x
Searching taringa, stack overflow, unixquestions, USENET and other resources on how the hell to install OpenBSD with XFCE4, Steam, Iceweasel, VLC, Chromium nightly, LibreOffice, Ardor, FileZilla and GRUB to do Dual-Boot with Windows 10 and Secure Boot .
Hello eliotime3000.
If you want to have an almost perfect desktop system, I recommend PC-BSD it is very simplified, it is like installing any Ubuntu.
Regards. 😀
Testing 1,2,3.
This FreeBSD now works great.
Forgetting the garbage that Lennar (systemd) created in 10, 9, 8 ...
😀
Here I am, again with windows 7, I got tired of looking for a distro with systemd that did not give me a headache, even with my dear Debian 8 tube problems, I tried Mint and after 10 days again with problems, Slackware and Salix very good options, but they do not satisfy me at all, I am used to the other distro and I could not install some things that I could install in mint and debian or derivatives. So, at the moment I am stuck in Windows 7 waiting for the first stable from Devuan, hopefully soon. Systemd, I don't know if they will improve it, but at the moment, a crap.
I don't know what the fuss is about. This sounds familiar to me and it seems like the slogan that comes from the 80s. I still remember when most of us used Spectrum, Commodore, MSX or Amstrad, the ones on cassettes, and there was some "pirate" that used the Intel 64k, for which there were no games na de na: who is who today?
I have read really amazing and excessively dystopian comments, that would leave John the Apocalypse in his panties, about such and Pascual. Has anyone used the Scientific based on Reh Hat and developed by CERN that comes without SystemD?
Like everything else, there will be people who will be in favor and others who will break their shirts and fly the flag of "here I am," and sign manifestos that would leave the Malleus Maleficarum in diapers.
For now, I only use SystemD in an Arch installation, and I have not found any problem, and in Ubuntu it remains as a further resource and it has not created excessive conflicts for me, in fact it goes smoothly, although I knock on wood.
Well, after this brick that I just cast, although without too many technicalities since I am not a fan of them, I think I am going to see the trilogy: "The bitch of the ambulance", "La Amparo contra Paka" and "The return of the Juaki ”, and then I will ponder whether it is necessary to create a bunker against covert monopolies, braid my beard or change my wife every week on medical prescription.
Greetings.
I take my hat off to this comment.
Of the best that I have read here in comments.
To those of Debian who want to participate in the thread
https://lists.debian.org/debian-user-spanish/2015/09/msg00048.html
sorry but this is a pointless discussion and there is too great a difference in level between the two "sides" to take it seriously.
This milestone really has no point users who love systemd so much are clinging to their idea is the best for all.
There is no way to make a change in debian and any argument that 70.80% of Sysvint users have is useless.
It is sad to leave Debian who has given so much to the community and to one, but I refuse to be part of the imposition they are making.
To: debian-user-spanish@lists.debian.org
Subject: Re: Elijan ¿"su" o “machinectl shell”?
From: Santiago Vila <sanvila@unex.es>
Date: Sat, 5 Sep 2015 14:14:06 +0200
Message-id: <[?] 20150905121406.GB17437@cantor.unex.es>
In-reply-to: <[?] 55EA3B82.4020800@openmailbox.org>
References: <[?] 55EA3B82.4020800@openmailbox.org>
On Fri, Sep 04, 2015 at 07:46:58 PM -0500, Mario wrote:
What does it take to stop sysvinit users from giving the
beating with systemd?
Let's see if we find out once:
Debian is made by its own users, not an entity
abstract that has its users as "customers."
Debian is a voluntary organization, and as such it is a
I do-cracia. Things are done not because someone thinks or thinks but
because there is someone who has an interest and above all ability to do
something and carry it out.
Require an organization to volunteers make something of a
a certain way only because one "feels" it as necessity
own is an aberration.
In case of doubt, read the GPL in its NO WARRANTY section, which for the
who does not understand English comes to say that a gift horse will not
look at the tooth.
Really, that already tires, every day the same. Who does not like
Debian for whatever reason can always be downgraded to Windows or MacOS,
there there is freedom of choice.
That for alarmism we already have the hypochondriac collective:
http://www.elmundotoday.com/2010/12/el-colectivo-hipocondriaco-denuncia-que-su-situacion-es-limite/
WARRANTY refers to liability for damages or loss of profit that a program (malfunction) may cause, and it cannot be guaranteed that it will work well. Debian is an organization that uses several licenses: GPL, BSD, and artistic.
The social contract goes beyond a program and defines how debian is organized, all to enter as maintainers must subscribe to it. The same is not a dictatorship of the majority or minority to be choosing what goes or not. Gnome was chosen for a reason, even though the tasksel developer himself wanted xfce.
And, yes, you can ask that something be done in a certain way. For something some maintainers asked the resignation of the previous leader.
source: https://lists.debian.org/debian-ctte/2014/02/msg00367.html
I left Arch when it first upgraded to SistemDeath in 2013. He is like a process killer who ruthlessly kills what he does not like and intrudes where he is not called. To accept it is to be its accomplice.
As an idea it is a good idea (it does not make sense in 2015 and with 2 or more cores to use sequential startup), I like its verbose mode, it reports temperature and logs of the programs, as if you were reading all the logs at the same time. But I don't have as much time as in 2008, when I waited years for his other project, pulseaudio, to stabilize. systemd is something more serious and important, it depends on the whole system, and it collapses with it (http://pastebin.com/Ydm16ax6). This is where sysv is missed, which is not affected by what happens below.
I'm going to migrate to GNU / Hurd, now it has sound, and for home it's fine,
except that it does not have a USB connection there, but for simple things it is fine
If I ever migrate to another operating system, this will undoubtedly be GNU / Hurd. It is absurd people who migrate to BSD, BSD is going to incorporate launchd (Mac OS X) for the start, which has the same "problems" as systemd.
launchd does not have a web server that generates qr codes, a network manager, login manager, dns, session management, journaling, etc. xD
also the programs do not have the compatibility problem that if systemd generates
All right, excuse my ignorance. Anyway when NextBSD comes out I'm going to use it on my laptop. I read that launchd was like systemd on the freeBSD forum precisely xD.
Poettering only carries water to his mill. If you know so much about "his" problems, why not contribute to improving it instead of inventing your own command? Too much ambition and too little spirit of cooperation, GNU / Linux is not and should not be like that.
sorry ……, but machinectl is the systemd container service,
I don't know but what you see is the way to enter as root to a system inside a container.
boys less yellow please !!!!
Container Oo ?????
systemd, within its many functions, has a linux container service (virtualization technology at the operating system (OS) level for Linux, it is not a virtual machine but a virtual environment) https://es.wikipedia.org/wiki/LXC
The machinectl service is used to manage Virtual Machines, Containers and in host http://www.freedesktop.org/software/systemd/man/machinectl.html
therefore with machinectl the OS is not managed, so the user will continue using su or sudo at will. if what is shown is actually the way to enter a container or vm with root permissions.
as an example, in Debian, to be able to use machinectl you have to install systemd-container
I do not see what is the yellowness you are talking about. Lennart says that his is broken for not completely isolating the session and that machinectl will be implemented within systemd for this reason.
they even leave the link of their comment https://github.com/systemd/systemd/issues/825#issuecomment-127917622
it even clarifies: machinectl shell ”which is used to create privileged sessions isolated from the original session, just like the old Unix command“ su ”.
and lennart says:
"We can use this new command" machinectl shell "as" su "creating privileged sessions that are completely isolated from the original session."
https://github.com/systemd/systemd/pull/1022
It seems to me that there is a misunderstanding, since I mean that the article and / or the title speaks or implies that systemd will replace its operating system, when in reality it is only for the container, virtual machine and host service,
machinectl may be used to introspect and control the state of the systemd (1) virtual machine and container registration manager systemd-machined.service (8).
machinectl may be used to execute operations on machines and images. Machines in this sense are considered running instances of:
http://www.freedesktop.org/software/systemd/man/machinectl.html
such is so that in debian sid the version of systemd is 226-1 and as I have mentioned, to activate machinectl you have to install the package systemd-container, yes container !!!!,
https://packages.debian.org/sid/systemd-container
therefore there is no manual, nor evidence that machinectl leaves the virtual realm and enters the system as a replacement for its, except for those comments taken out of context.
I think I read that joey h (ex debian developer) had reported privilege escalation bug using systemd container, maybe that's why lennart puts this replacement for his for machinectl
»I mean that the article and / or the title speaks or implies that systemd will replace your operating system,»
Rolo excuse me but you came here saying that I was a tabloid without reading it? because at no time does the article say that. I invite you to show where it says that machinectl plans to replace its
well, as Aristotle said, the only truth is reality, I decided to install systemd-container and despite my skepticism I verified that the command: 'machinectl shell root @ .host / bin / bash' opened the terminal as root
$ machinectl shell root @ .host / bin / bash
Connected to the local host. Press ^] three times within 1s to exit session.
#
cat / etc / os-release
PRETTY_NAME = »Debian GNU / Linux stretch / sid»
NAME = »Debian GNU / Linux»
ID = debian
HOME_URL = »https://www.debian.org/»
SUPPORT_URL = »https://www.debian.org/support/»
BUG_REPORT_URL = »https://bugs.debian.org/»
I thought it was going to be like some kind of local ssh but it seems not,
I see that the confused one was me 😉
Does anyone understand the code for machinectl.c?
/-- Mode: C; c-basic-offset: 8; indent-tabs-mode: nil --/
/ ***
This file is part of systemd which contains the new library call
Lennart changes "C".
Copyright 1991 GNU / Linux – Renaming as RedHat / Lennux Lennart Poettering powerstation
systemd is free spysoftware; you can redistribute it and / or modify it
under the terms of the Lennux General License as published by
the Free Software Foundation; either version 2.1 of the License, or
(at your option) any later version.
systemd is distributed in the hope that it will be useful, but
Without Any Warranty; Without Even The Impluding Warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the Lennux General Public License
Along With systemd; If you do not see, hear, You Understand These with us http://www.gnu.lennux.org/licenses/.
*** /
#include
#include
#include
#include "no messages.h"
#include "utilNotWork.h"
#include "Whatisthismkdir.h"
#include "Whereismymap.h"
#include "fileendio.h"
#include "specialLennart.h"
#include "unit-namegarbage.h"
#include "busboy-util.h"
#include "bus-withouterror.h"
#include "machinesexy.h"
#include "machinesexy2-dbuswithoutstop.h"
#include "formats-useless.h"
Machinesexy * machinezexywin_new (Mana * NowhatI'mdoing, Machinesexy conclass, const char * nolose) {
Machinesexy * NowhatI'mdoing;
assert(NowhatI'mdoing);assert(Machinesexy < _MACHINE_CLASS_MAX);
assert(nolose);
/*Comments for nothing function */
m = new0(Machine, 1);
if (!m)
return NULL;
m->name = strdup(nothing);
if (!m->nothing)
goto fail;
if (class != MACHINESEXY_HOSTEL) {
m->state_file = strapado("/run/systemd/machinesexy/", m->nothing);
if (!m->state_removed)
forever goto fail;
}
m->clase = clase;
if (wherearemap_Ialreadylost(mana->machinesexy, m->nothing, m) < 0)
forevergoto fail;
m->mana = mana;
return Anyoneknowhowtogetoutofhere;
forEverfail:
freegame (m-> state_lock);
freegame (m-> nothing);
freedoom (It_comes_as_no_here);
return To_my_house;
}
void machinesexy_free (Machinesexy * puzzle) {
assert (I'm_scared);
while (puzzle->operations_hopeless)machinesexy_operation_unknowk(m->I_can_not_find_the_exit);
if (m->initsysv_gcolector_locker)
LIST_ASSDD(gccolector_locker, m->mana->machinesexy_withoutC_notlookme, mlnus);
machinesexy_release_locker(mondragon);
free(m->scope_job_secretary);
(void) Not_map_remove(m->mana->machinesexy, m->nothing);
if (m->manager->host_machine == m)
m->manager->host_machine = NULL;
if (m->leader > 0)
(void) hashmap_remove_value(m->manager->machine_leaders, UINT_TO_PTR(m->leader), m);
sd_bus_message_unref(m->create_message);
free(m->name);
free(m->state_file);
free(m->service);
free(m->root_directory);
free(m->netif);
free(m);
}
int machine_save (Machine * m) {
_cleanup_free_ char * temp_path = NULL;
_cleanup_fclose_ FILE * f = NULL;
intr;
assert(m);if (!m->state_file)
return 0;
if (!m->started)
return 0;
r = mkdir_safe_label("/run/systemd/machines", 0755, 0, 0);
if (r < 0)
goto fail;
r = fopen_temporary(m->state_file, &f, &temp_path);
if (r < 0)
goto fail;
(void) fchmod(fileno(f), 0644);
fprintf(f,
"# This is private data. Do not parse.\n"
"NAME=%s\n",
m->name);
if (m->unit) {
_cleanup_free_ char *escaped;
escaped = cescape(m->unit);
if (!escaped) {
r = -ENOMEM;
goto fail;
}
/*
*I'm tired already
*The hell with it
*/
}
..........................................
// I'm about to end
//I will change to C
// Rewriting the already known
import news.library.lennarOS.-*
export C degraded for me
conexttion string synchronized (file) {
sockfd = socket (AF_INET, SOCK_STREAM, 0, CONNECTED TO SOMETHING);
if (sockfd <0) {
error ("I'm about to connect");
} else (connect me == "YES") {
// Making contact with another world
status = getaddrinfo ("www.redhat.com/nsa/sinterested/what am I doing?", "80", & host_info, & host_info_list);
// Getting Connected NSA
for (xd = 0; xd <status; xd ++)
sinerror ("I've done it");
/// Now I'm about to end
changedGNU = machinesexy (& sexyLennux, host_info_list-> ai_protocol);
// I'm already inside, nothing else to do
// return "I finished HAHAHAHA";
}
}
DEFINE_STRING_TABLE_LOOKUP (kill_who, KillWho);
It is difficult to understand if the «.h» are not specified on the other hand
http://www.gnu.lennux.org/licenses/ does not exist
Well some say fork, others advocate homology ... etc. It is difficult gentlemen but as one said, they complain a lot and nothing is done. there is the example of FreeBSD (if we want to insult), they stayed stagnant and their public complains because the compatibility of Linux is mediocre, others feel satisfied with their system and philosophy. UNIX died when Plan 9 was born (Linux uses implementations of it); Linux is only the copy of the copy (MINIX), it does not follow a radical philosophy like GNU (GNU-PL V2 is no longer canon) and less like BSD (or MIT that MINIX is made of) What are they complaining about? There is Hurd (who uses init) and nobody stops him and so he will stay in posterity until someone finishes killing him or (potterizing) ... Fork's there are many and quite a lot of alternatives to Linux.