SoylentNews is people
SoylentNews
Longtime Debian contributor Tollef Fog Heen has announced his resignation from the Debian systemd maintainer team. His announcement states that "the load of the continued attacks is just becoming too much."
He has since written a detailed blog article surrounding the circumstances of his resignation. As he puts it,
I've been a DD for almost 14 years, I should be able to weather any storm, shouldn't I? It turns out that no, the mountain does get worn down by the rain. It's not a single hurtful comment here and there. There's a constant drum about this all being some sort of conspiracy and there are sometimes flares where people wish people involved in systemd would be run over by a bus or just accusations of incompetence.
This is yet another dramatic event affecting the Debian project in recent months. The adoption of systemd has been extremely controversial, even going so far as to result in calls for Debian to be forked. There have been other problems as of late, too, ranging from a serious bug breaking Wine just days before the Jessie freeze deadline, to the possibility of Debian GNU/kFreeBSD being dropped from Debian 8. And it was only just over a week ago that Joey Hess — another longtime Debian contributor — left the project, citing the "very unhealthy directions" that Debian has been led in lately.
Is the internal tension and strife caused by systemd about to tear the Debian project apart? Recent events such as the aforementioned have suggested that this is becoming more and more of a possibility. The repercussions of this drama will no doubt be felt wide and far, given Debian's own popularity, as well it forming the basis of other major Linux distros such as Ubuntu and Linux Mint.
(Score: 2, Insightful) by Anonymous Coward on Monday November 17 2014, @01:59PM
Major distros are using it because the legacy init system was antiquated and had become a pain in the neck: among other complaints, there is no good and reliable way to express dependencies between system services so every startup script had to do its own checking to see if it had the prerequisites to start. Package maintainers didn't like it.
It was obvious that an improved init infrastructure was needed. The controversy, as I see it, is primarily about whether the cure is worse than the disease, and secondarily about who did what with whose mother.
(Score: 2, Interesting) by jbruchon on Monday November 17 2014, @03:06PM
What Linux (and all UNIX-like systems) really needs is the equivalent of Microsoft's Service Control Manager. Everything in Linux/UNIX is a kludge by comparison; systemd uses cgroups to corral services, but there's no way for systemd to actively communicate with and poll the status of a program like Apache or dnsmasq so that they can be properly managed. Most init systems don't monitor services at all; a totally different (non-standard, since service management on UNIX-likes is the Wild West) program tends to be set up to handle this. There are no standards for "system services," they are absolutely indistinguishable from normal executables.
We need a standardized, minimal, and **very well documented** service control and monitoring C API that allows programs to interface with (and be polled by) an arbitrary "service control manager" on Linux. This could be done and not violate the UNIX philosophy of "do one thing and do it well." It would also fix the main problem that systemd was initially created to solve: kludgey "reinvent the wheel for each program and hope no PIDs sneak away" management of system services.
I'm just here to listen to the latest song about butts.
(Score: 0) by Anonymous Coward on Monday November 17 2014, @04:47PM
Huh? Are you for real?
Sysvinit isn't perfect, but I'd take it over the Windows approach any day. I say this as someone who has managed Windows systems since NT 4, along with all sorts of Linux, BSD and UNIX systems.
(Score: 1) by jbruchon on Tuesday November 18 2014, @06:15AM
I'm not advocating "the Windows approach." I'm suggesting that there be a standardized way for system services to be brought up and down and polled for lockups. I'm not suggesting that a service manager be the ONLY way to start services, either.
I'm just here to listen to the latest song about butts.
(Score: 0) by Anonymous Coward on Tuesday November 18 2014, @12:49PM
Let me quote what you wrote:
What Linux (and all UNIX-like systems) really needs is the equivalent of Microsoft's Service Control Manager.
You are advocating the Windows approach.
(Score: 1) by jbruchon on Tuesday November 18 2014, @01:11PM
By your logic, saying Nautilus is "the equivalent of Explorer from Microsoft Windows" is false. "Equivalent" does not mean "identical."
I'm just here to listen to the latest song about butts.
(Score: 2) by TheLink on Monday November 17 2014, @05:07PM
What?!! Pre-Vista stuff was like this: https://support.microsoft.com/kb/203878 [microsoft.com]
e.g. services have dependencies NO Windows doesn't check them during shutdown!
It took them so long, and then they implement it with preshutdown notifications AND that's not supported for .Net!:
http://msdn.microsoft.com/en-us/library/windows/desktop/ms685149%28v=vs.85%29.aspx [microsoft.com]
http://connect.microsoft.com/VisualStudio/feedback/details/641737/add-windows-service-preshutdown [microsoft.com]
See also: https://stackoverflow.com/questions/7437590/is-it-possible-to-register-for-preshutdown-service-events-using-net [stackoverflow.com]
Microsoft could in theory have created a C++ service that handles preshutdowns of .Net stuff. Maybe one day I'll do it if nobody else gets around to doing it (I'm not a great coder - so I'd rather not unleash another crappy C++ service on the world if possible).
(Score: 1) by jbruchon on Tuesday November 18 2014, @01:20PM
If someone wrote a "service manager" for Linux, none of what you just wrote would matter. No one suggested "porting Microsoft's services.exe program" here. Everything that attempts to improve on or replace sysvinit tries to manage services better; I'm saying that we need services to be able to talk back to the program that manages them and that program needs to be able to do things like poll the services to make sure they haven't done something bad, such as having crashed without the PID terminating.
I'm just here to listen to the latest song about butts.
(Score: 2) by forsythe on Monday November 17 2014, @05:13PM
there's no way for [an init system] to actively communicate with and poll the status of a program like Apache or dnsmasq so that they can be properly managed
I hope I'm misreading you, because if you're saying what I think you're saying, then you'll have to explain to `rc-status` that it doesn't exist.
(Score: 0) by Anonymous Coward on Monday November 17 2014, @11:25PM
He's saying what you think he's saying. And, yes, he's wrong. Pretty much everything in his comment is dumb. Seriously, he's suggesting using the approach that Windows takes, including its well-known problems, and bringing it over to Linux. That's really fucking stupid thing to suggest.
(Score: 1) by jbruchon on Tuesday November 18 2014, @06:02AM
Says someone who obviously had no better suggestions.
I'm just here to listen to the latest song about butts.
(Score: 0) by Anonymous Coward on Tuesday November 18 2014, @12:57PM
We don't have to suggest anything new. Sysvinit has worked perfectly fine for a long time. We don't need a "better suggestion" because nothing needs to change.
(Score: 1) by jbruchon on Tuesday November 18 2014, @01:13PM
I disagree. I have been using runit for a long time and it's definitely better than sysvinit. http://busybox.net/~vda/init_vs_runsv.html [busybox.net]
I'm just here to listen to the latest song about butts.
(Score: 1) by jbruchon on Tuesday November 18 2014, @06:12AM
http://dev.gentoo.org/~vapier/openrc/projects/openrc/ticket/120.html [gentoo.org]
http://dev.gentoo.org/~vapier/openrc/projects/openrc/ticket/145.html [gentoo.org]
Also, how can OpenRC know that e.g. httpd hit a bug and is hosed but the httpd service PID has not terminated?
I'm just here to listen to the latest song about butts.
(Score: 2) by cafebabe on Tuesday November 18 2014, @05:56AM
I thought that was #gamergate.
1702845791×2