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: 3, Insightful) by jmorris on Monday November 17 2014, @04:54PM
I do think there is a distinct stench of astroturf but it isn't the primary issue. The problem is there are two arguments running in parallel and co-mingling.
Is systemd a viable technical project and does it actually work or is it likely to in the near future?
All of the arguing about bugs and Pottering's history of leaving smoking wreckage in his wake are tied to this question. It is not the most interesting of the arguments. Bugs can be fixed, Pottering will move on to wrecking something else eventually. So if you agree with the goal, the technical issues revolve more around whether systemd is ready yet, so even the most extreme views against systemd in this camp would say to wait a few years and let it prove itself but it is only a question of when the change happens.
The other argument is over whether systemd is a good idea in the first place.
By systemd I mean the whole Pottering "UNIX Hater's Club" project of systemd as an OS in userspace, absorbing everything in its path. It means the adoption of an overly complex port of Service Manager and Event Logger, burying everything under layers of virtualization and sandboxing to get security instead of fixing the damned bugs in the first place at the price of making the internals so complex that new admins will require years to understand what is actually happening instead of a month or two.. which will have its own security implications. It means objecting to the whole idea of tossing POSIX and the essentially the whole gnu-utils package and the other UNIX classics like grep, etc. In short, do we want a free Windows or a UNIX?
Note that this is a cultural question. This is why the Debian TC had so most strife over it and their final verdict had almost zero impact on the NO side. Because they were trying to solve a problem outside their mandate.
You can't compromise on this issue, you can split the difference. It is a fork in the road, taking one precludes the possibility of the other. There is going to be a fork, the only question is whether enough people realize it in time to avoid a lot more developers getting burned out by the fighting.
(Score: 2, Informative) by fritsd on Monday November 17 2014, @06:00PM
Very well put; I think the crux is whether systemd is a good idea in the first place (your second argument).
It seems to me that systemd is written by geniuses, but not written *for* common-as-muck sysadmins who use Linux in production, instead for some mythical average desktop use case.
Try to read some of the comments on http://debianfork.org/ [debianfork.org] (scroll down; about 50% rants, 50% complaints of people who sound like real actual sysadmins, and a website that's easy to read for the 60+ sysadmin with poor eyesight)
I think the best comment there (can't attribute; the author is not shown) is:
"There is a debate whether to replace legacy init-systems. It is a good
debate, and imho a new init system is very due.
What should have been done (*):
1.) define interfaces/apis for a new init system by the linux
community/process
2.) standardaize these interface
3.) have somebody provide a reference implementation and
reference-test-suite (an init-system is missing critical, I cannot debug
umteenth servers when they fail initing)
what has been done:
1.) a reference implementation has been pooped into existence with
interfaces/apis 'designed' on the fly
2.) this mix of standards/implemention has then been pushed and force-fed
to the community
3.) now the community is pissed
The discussion about whether or not systemd must be used is moot. If
standards exist, systemd can be replaced. If not, like we have now, it
cannot."
(Score: 2) by melikamp on Monday November 17 2014, @08:50PM
(Score: 0) by Anonymous Coward on Tuesday November 18 2014, @03:59AM
It is just a bunch of nastiness.
Of course it is "viable" it is already in widespread use. Most linux users are running it already, without any problems. Slandering the author is silly, his past projects that get name-called are the ones that won on technical merits, fixed the bugs, and are still in use. Still in use. Still in use. That is not failed. That is succeeded. D'oh!
(Score: 1) by jmorris on Tuesday November 18 2014, @07:37AM
No. I'm typing this on a Thinkpad running F20. PulseAudio is f*ck*d. It has been f*ck*d for years and years on pretty much every platform I have had the misfortune to use it.... which because RH and the GNOMEs rammed it down everyone's throat is almost everywhere; exactly like systemd. Years have passed and I'm sick and damned tired of listening to excuses for that useless piece of crap that you can't get rid of without a week of intense effort. Just because someone might have a bluetooth headset and want to 'seamlessly' (yea, that will be the day) transition the stream between their headset and the speakers we have all had to suffer broken audio for almost a decade now.
Drop onto the docking station, a couple of RANDOM outputs will mute themselves. Suspend and guess what happens? Undock and guess what happens. Launch the PulseAudio Graphic EQ plugin and GUESS WHAT HAPPENS! While you could attribute the dock and power management related ones to potential kernel bugs, the EQ doing exactly the same thing kinda gives the game away.
Better, since I ditched Gnome3 for Mate there isn't even a graphical tool remaining (unsure if there is one in Gnome but there migh) that can even manipulate the actual hardware volume controls, only pulse. All I have found is alsamixer in a terminal window. (Although the GUI alsamixer would be viable too with the -c0 switch... since it has no graphical way to select the card to control and defaults to pulse.)
No, I will never allow Pulse, Networkmanager or systemd near a production server. Never. When Debian becomes unusable I'll look for a fork. If there isn't one Slackware. Should it too fall to the forces of darkness BSD will always be there.
Because I'm in both dissenting camps. systemd is a bad idea that isn't likely to ever work reliably and securely. Or at least not work in the twenty or so years I have left before retirement.