The Switch To Systemd Will Likely Occur For Ubuntu 15.04

[Luke]

Much of that an maintainers will only deal with source

Ironic is a gentle word to use in such case.
I have a question to the team behind the last GR and its real purpose:
how you think to manage the situation where the maintainer of the package X won't maintain all the init system of the universe at the cost of *his* free time and *his* work, instead of the work and time from who want a different init system to be supported?
What if a lot of devs start to orphan their packages? What's happen then?
I'm sure Ian will be able to destroy debian, let him to complete his mission under the name of "the freedom of the project's suicide".

A project maintainer could deal with this by providing only source, supporting only systemd, and simply advising anyone building debian packages to mark them in DEBIAN/control as depending on systemd. Alternately, if the mantainer is a Debian user, they could publish the debian packages from their own repo done their way, no need to abandon it.

There are plenty of packages in my Ubuntu-based systems that don't come from official repos. Kdenlive, MLT, Mesa, and the kernel are all from PPA's, 15 are my own debs made from builds of ffmpeg from source (needed for Kdenlive to work right with AVCHD files!), one is a repackaged and modified version of Debian's dracut package, and ten are my own programs outright.

Debian is free software and so are its downstream derivatives like Ubuntu and it's derivatives like Mint. Packaging policies only affect those trying to get into a distro's official or "community supported" default repos. They don't deem themselves to have the authority to prevent users from installing other packages.

[bison]

So I have until April 2015 to find an alternative to Ubuntu.

It may turn out to be not that big of a deal. There are quite a few people who don't want to switch to systemd, so there will almost certainly be alternatives. There might even be a Ubuntu variant that continues with upstart. So far all the Ubuntu "flavours" have had alternate desktops, but there's no reason that one couldn't have an alternate init system. There would be a few things you couldn't run, such as Gnome, but that's not likely to be a big deal since people who dislike systemd tend not to be Gnome users.

[revellion]

I don't get it

Why is there such hatred for SystemD?.

I was initially a bit sceptical of it aswell being used to sysvinit.

But after having used SystemD for almost a year now, it's made stuff a lot more simpler and not to mention consistent.

Before if i was to advice someone how to autostart a service on LInux or how to start one. I'd have to think of which of the bastardly different ways each distro did it before.

Now it's simply

systemctl enable someservice

or

systemctl start someservice

rather than

service someservice start
or
/etc/init.d/someservice start

or some other odd way on Red Hat/Fedora.

[Pajn]

Why is there such hatred for SystemD?.

It's eating everything. If you don't want to use SystemD you have a very small
choice of good quality distros and can't use a pretty huge clunk of programs.

Technically there isn't anything wrong with it but it create a few issues. For example,
if there is a bug in SystemD then the whole Linux world is compromised.

Personally I will stick with Ubuntu and thus soon start to use SystemD, however
I'm glad to be on the last big distro that makes the switch. Not just because of
the above but also that I expect you guys to have found and fixed all the major
bugs for me

[Nobu]

Technically there isn't anything wrong with it but it create a few issues. For example,
if there is a bug in SystemD then the whole Linux world is compromised.

But is that really such a big deal? If that bug is patched, the whole Linux world is saved! On a more serious note: you probably aren't using multiple distributions, on various computers, simultaneously. Even if you are, are you going to switch distributions because of an init bug? Most people would either wait for it to be patched and sent through the update channels, or patch/update themselves. If it's a critical bug (iow, a crasher or some-such), you'd probably chroot in and roll-back or update to a fixed version, just like with any other init, though that kind of bug is usually caught in testing before it's rolled out to stable.

I know I didn't switch from Ubuntu to Fedora (and finally Arch) just because Upstart was an unstable mess when it was first being tested. I was anxious, and a bit off-put, but it was a combination of things which ultimately drove me away (I didn't really care for the direction Ubuntu as a whole was going, Unity wasn't really stable with my system configuration, and I was more comfortable with a rolling release model, though I didn't really settle on that until I had tried a few other distros in VMs and on hw).

[Pajn]

But is that really such a big deal? If that bug is patched, the whole Linux world is saved! On a more serious note: you probably aren't using multiple distributions, on various computers, simultaneously. Even if you are, are you going to switch distributions because of an init bug? Most people would either wait for it to be patched and sent through the update channels, or patch/update themselves. If it's a critical bug (iow, a crasher or some-such), you'd probably chroot in and roll-back or update to a fixed version, just like with any other init, though that kind of bug is usually caught in testing before it's rolled out to stable.

I know I didn't switch from Ubuntu to Fedora (and finally Arch) just because Upstart was an unstable mess when it was first being tested. I was anxious, and a bit off-put, but it was a combination of things which ultimately drove me away (I didn't really care for the direction Ubuntu as a whole was going, Unity wasn't really stable with my system configuration, and I was more comfortable with a rolling release model, though I didn't really settle on that until I had tried a few other distros in VMs and on hw).

As Linux is used on a majority of the worlds servers a vulnerability that affects them all have an extremely high price. Previously they haven't shared so much that just a single vulnerability could affect them all (the kernel is usually so differently configured that hacks on it aren't very portable). Now they will share SystemD so hacking SystemD suddenly became very interesting.

Sure updates will come real fast as everyone is in the same boat, but to gain from that I need to be on patch-watch 24/7.

[finalzone]

As Linux is used on a majority of the worlds servers a vulnerability that affects them all have an extremely high price. Previously they haven't shared so much that just a single vulnerability could affect them all (the kernel is usually so differently configured that hacks on it aren't very portable). Now they will share SystemD so hacking SystemD suddenly became very interesting.

Please call by proper name: systemd

Yes, it is written systemd, not system D or System D, or even SystemD. And it isn't system d either. Why? Because it's a system daemon, and under Unix/Linux those are in lower case, and get suffixed with a lower case d. And since systemd manages the system, it's called systemd. It's that simple. But then again, if all that appears too simple to you, call it (but never spell it!) System Five Hundred since D is the roman numeral for 500 (this also clarifies the relation to System V, right?). The only situation where we find it OK to use an uppercase letter in the name (but don't like it either) is if you start a sentence with systemd. On high holidays you may also spell it s?st?md. But then again, Syst?me D is not an acceptable spelling and something completely different (though kinda fitting).

[gens]

well it's gonna reach version 500 soon anyway

[Nobu]

They also use Apache, sshd, and various other standard tools (bash, for instance), and most of those, arguably, have a larger attack surface. Not every server out there is running the same version of systemd, either (not to mention the configuration, which is easily changed and could differ as much as anything else). It's probably even more varied than the other init's out there, simply because of how often new versions of systemd are released.

[valeriodean]

A project maintainer could deal with this by providing only source, supporting only systemd, and simply advising anyone building debian packages to mark them in DEBIAN/control as depending on systemd. Alternately, if the mantainer is a Debian user, they could publish the debian packages from their own repo done their way, no need to abandon it.

I have hard time to imagine a debian official repo full of source files only, but even in that case, the GR's proposal say more that that, it forces the maintainer to accept any path that add a compatibility with a inist systems other that the official one.
But from that point onwards, the maintenance burden for that package is on the maintainer shoulder, no more on who proposed the patch.
So the maintainer should run every init system just to be sure that the code continue to work for all the next releases for all the init systems?
Isn't the role of a maintainer to be sure that the package continue to run smoothly in every release after every update?
Then how can you say a project maintainer could deal with this by providing only source, supporting only systemd?