Gentoo guys starting a fork of udev

[Thomas Goirand]

Hi,

I think this is an interesting read:
http://article.gmane.org/gmane.linux.gentoo.project/2262

The full thread is here:
http://thread.gmane.org/gmane.linux.gentoo.project/2262

As Gentoo guys and some major kernel people are protesting about the
insanity Kay and Lennart have done to udev, probably this is a window of
opportunity to stop being annoyed in Debian, by brutal RedHat decisions
which we don't necessarily (have to) agree on (eg: move to /usr, merge
with systemd, etc.).

Thoughts anyone? Would anyone have time to work with Gentoo? What is the
view of the kernel team and the udev (marco?) maintainers?

Cheers,

Thomas


--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org
Archive: http://lists.debian.org/50A30C23...@debian.org

[Michael Biebl]

On 14.11.2012 04:12, Thomas Goirand wrote:
> Hi,
>
> I think this is an interesting read:
> http://article.gmane.org/gmane.linux.gentoo.project/2262

So far this seems to be mostly talk and hot air.
I'm also wondering how many more forks the gentoo guys want to start [1].

Michael


[1]
https://forums.gentoo.org/viewtopic-t-934678-postdays-0-postorder-asc-start-0.html

--
Why is it that all of the instruments seeking intelligent life in the
universe are pointed away from Earth?

[Thomas Goirand]

On 11/14/2012 11:25 AM, Michael Biebl wrote:
> So far this seems to be mostly talk and hot air.

It's clearly going to take some time to materialize into a more
definitive project, however, I don't think that's fair to say it's
only "talk and hot air" as I saw some Gentoo patches to
uncruft udev already. Though the thread I gave link to doesn't
show that at all.

Thomas


--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/50A352F4...@debian.org

[martin f krafft]

also sprach Thomas Goirand <zi...@debian.org> [2012.11.14.0412 +0100]:

> As Gentoo guys and some major kernel people are protesting about the
> insanity Kay and Lennart have done to udev,

I cannot help but notice that Kay and Lennart were both
Gentoo-freaks when they took on udev and at least I always
attributed much of what was wrong with udev from the start (e.g. the
configuration file format) to being born in an environment where
people still compile from source. ;)

--
.''`. martin f. krafft <madduck@d.o> Related projects:
: :' : proud Debian developer http://debiansystem.info
`. `'` http://people.debian.org/~madduck http://vcs-pkg.org
`- Debian - when you have better things to do than fixing systems

"convictions are more dangerous enemies of truth than lies."
- friedrich nietzsche

[Matthias Klumpp]

What are the problems they try to address? The strong binding to
systemd is good and makes much sense to me, and udev is still usable
without systemd (and will be in the future).
Also, both systemd and udev are Linux-only, so the situation here at
Debian hasn't changed.
The problems we had in the past with bad udev+kernel combinations and
changing config file format etc. can also be addressed in udev,
without the need of forking.
In general, I think a fork of udev would do much more harm than trying
to solve the problems in udev. Of course, they're free to fork, but
the separation will hurt both projects and everything relying on
udev/the fork.
Regards,
Matthias

2012/11/14 martin f krafft <mad...@debian.org>:

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/CAKNHny_vVFoQziF1H4oAVWZZ...@mail.gmail.com

[Patrick Lauer]

On 11/14/12 19:53, Matthias Klumpp wrote:
> What are the problems they try to address?

Removal of features, broken build system, etc.
Just the little things that make things exciting (and we prefer things
to be boring so we can sleep at night)

Plus an unpredictable upstream that can't be trusted anymore.
(See recent rants by kernel people about udev breaking firmware loading
completely because LOL FASTOR)

> The strong binding to
> systemd is good and makes much sense to me,

If you wanted to use systemd maybe. But since we don't want it it's a
strong negative point

> and udev is still usable
> without systemd (and will be in the future).

It's slowly losing features or getting broken in funny ways because
upstream wants you to migrate into the exciting future. It's becoming
more and more troublesome, and we don't know what they'll change next
just because they can

> Also, both systemd and udev are Linux-only, so the situation here at
> Debian hasn't changed.
> The problems we had in the past with bad udev+kernel combinations and
> changing config file format etc. can also be addressed in udev,
> without the need of forking.

Most of the issues we've had since the merge have been declared features
by upstream. Discussing with them appears to be futile.

> In general, I think a fork of udev would do much more harm than trying
> to solve the problems in udev. Of course, they're free to fork, but
> the separation will hurt both projects and everything relying on
> udev/the fork.

An API-compatible fork should not cause any problems. Since we cannot
cooperate with upstream I don't see any other way forward. I do agree
that it's "stupid" - would be nice if we could work together etc. etc.


> Regards,
> Matthias

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/50A3A387...@gentoo.org

[Patrick Lauer]

On 11/14/12 18:37, martin f krafft wrote:
> also sprach Thomas Goirand <zi...@debian.org> [2012.11.14.0412 +0100]:
>> As Gentoo guys and some major kernel people are protesting about the
>> insanity Kay and Lennart have done to udev,
>
> I cannot help but notice that Kay and Lennart were both
> Gentoo-freaks when they took on udev

You make my head hurt. These guys are throwing mud at Gentoo at any
chance they get, how on earth did you get an impression that they'd
consider Gentoo to be more than a kinky toy for bored kiddies?

> and at least I always
> attributed much of what was wrong with udev from the start (e.g. the
> configuration file format) to being born in an environment where
> people still compile from source. ;)
>

And you compile from what? ;)

But anyway, we're getting tired of their ADHD-driven changes just to
change things - maybe we can build up enough momentum so that things
might just be less frustrating for us all. You're all welcome to join,
ignore us or do what you want.

Have a nice day,

Patrick

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/50A3A153...@gentoo.org

[John Paul Adrian Glaubitz]

On Wed, Nov 14, 2012 at 09:49:07PM +0800, Patrick Lauer wrote:
> But anyway, we're getting tired of their ADHD-driven changes just to
> change things

TBH, I'm getting tired of people who are constantly shooting against
them because these people are unwilling to accept changes. We're not
bringing Linux forward if we stick to 30-year-old concepts. systemd is a good
design and most people actually agree otherwise it wouldn't become
standard on so many distributions (except Ubuntu, but that's rather a
political decision IMHO).

One of the Arch developers actually made a couple of good points why
they switched to systemd as default [1].

Cheers,

Adrian

> [1] https://bbs.archlinux.org/viewtopic.php?pid=1149530#p1149530

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121114140...@physik.fu-berlin.de

[Patrick Lauer]

On 11/14/12 22:04, John Paul Adrian Glaubitz wrote:
> On Wed, Nov 14, 2012 at 09:49:07PM +0800, Patrick Lauer wrote:
>> But anyway, we're getting tired of their ADHD-driven changes just to
>> change things
>
> TBH, I'm getting tired of people who are constantly shooting against
> them because these people are unwilling to accept changes. We're not
> bringing Linux forward if we stick to 30-year-old concepts.

Y'know ... I like progress. There's some pretty nice new things that are
an actual improvement, like GPT. At last more than 4 partitions (and no
longer compatible to MS-DOS 3.3) ...

But then there's for example UEFI (which I've never seen working as
suggested by documentation), grub2 (should be called "blinky cursor
app"), then the random changes to udev that make it unable to load
firmware, moving things around (well, no script hardcodes /sbin/ip,
right?) just for fun ...

I'm tired of these changes that don't solve any problems. Half-baked
stuff that is deployed before it is even feature-complete with the
boring old stuff it is supposed to replace. How would you feel about a
forced upgrade of apt to yum? After all newer is better ...

> systemd is a good
> design and most people actually agree otherwise it wouldn't become
> standard on so many distributions (except Ubuntu, but that's rather a
> political decision IMHO).

It does have some good ideas, and it is better than the random bits of
unmaintained shell it replaces - but it's mediocre at best. No real
design, just things nailed together with screws and secured with tape.

I'm waiting for the one that comes to replace it :)

> One of the Arch developers actually made a couple of good points why
> they switched to systemd as default [1].
>

Their users really appreciate it, especially those that are now
migrating to other distros because they preferred their OS when it was
booting as intended.

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/50A3AA92...@gentoo.org

[The Wanderer]

On 11/14/2012 09:04 AM, John Paul Adrian Glaubitz wrote:

> On Wed, Nov 14, 2012 at 09:49:07PM +0800, Patrick Lauer wrote:
>
>> But anyway, we're getting tired of their ADHD-driven changes just to change
>> things
>
> TBH, I'm getting tired of people who are constantly shooting against them
> because these people are unwilling to accept changes. We're not bringing
> Linux forward if we stick to 30-year-old concepts.

While I think I recognize the idea behind this, and it's one I don't necessarily
entirely disagree with, this sentiment just seems wrong to me.

True, if people are unwilling to accept change for the simple reason that it
*is* change, that's a bad thing.

But why is a 30-year-old concept necessarily worse than a new one? Or to put it
another way, why is it necessary to "bring Linux forward", in cases where what
is already present is good and works well? (And, taken further: in cases where
what is already there *isn't* good and/or *doesn't* work well, why is it
necessary to accept change *in a particular direction*, if that direction has
problems of its own?)

I've run across a few software projects where it has seemed as if the developers
were adding new features and removing old ones and changing UIs not because
there was something wrong with the old, but apparently just because "we're the
developers, we have to make changes or we're not developing it" - because they
seemed to think that letting a program sit unchanged is automatically a bad
thing, no matter how close to perfect-for-its-purpose the program may already
have been.

Change is not always bad; in fact, it's very often good. But change isn't always
good either, and refusal of change isn't always simple obstinacy or stagnation.

--
The Wanderer

Warning: Simply because I argue an issue does not mean I agree with any
side of it.

Every time you let somebody set a limit they start moving it.
- LiveJournal user antonia_tiger

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/50A3A895...@fastmail.fm

[John Paul Adrian Glaubitz]

On Wed, Nov 14, 2012 at 09:20:05AM -0500, The Wanderer wrote:
> But why is a 30-year-old concept necessarily worse than a new one? Or to put it
> another way, why is it necessary to "bring Linux forward", in cases where what
> is already present is good and works well? (And, taken further: in cases where
> what is already there *isn't* good and/or *doesn't* work well, why is it
> necessary to accept change *in a particular direction*, if that direction has
> problems of its own?)

Because System V Init isn't a good concept. It fails in so many
regards. There is no standardized way for init scripts, it cannot
make sure processes actually run and restart them on demand. It also
lacks mechanisms for ressource control and figuring out dependencies
between service without hardcoding them. It's just a dirty
hack. System V Init was good 20 years ago, but it isn't nowadays.

Automatic dependencies, process watchdogs and ressource control are
something which is incredibly useful to have nowadays, especially on
big machines which are shared among many users (clusters, for
example).

> I've run across a few software projects where it has seemed as if the developers
> were adding new features and removing old ones and changing UIs not because
> there was something wrong with the old, but apparently just because "we're the
> developers, we have to make changes or we're not developing it" - because they
> seemed to think that letting a program sit unchanged is automatically a bad
> thing, no matter how close to perfect-for-its-purpose the program may already
> have been.

True, but as I said, System V Init is not a good concept anymore,
that's why it's being dropped. Apple dropped the old init system with
MacOS X 10.4, why should the Linux world still stick to it in 2012?

Adrian

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121114143...@physik.fu-berlin.de

[John Paul Adrian Glaubitz]

On Wed, Nov 14, 2012 at 10:28:34PM +0800, Patrick Lauer wrote:

> I'm tired of these changes that don't solve any problems. Half-baked
> stuff that is deployed before it is even feature-complete with the
> boring old stuff it is supposed to replace. How would you feel about a
> forced upgrade of apt to yum? After all newer is better ...

Comparing apples and oranges here. You cannot "upgrade" apt to yum
because yum is feature-wise very much comparable to apt. systemd,
OTOH, brings many new useful features you simply can't have with any
other init system. Period.

>
> > systemd is a good
> > design and most people actually agree otherwise it wouldn't become
> > standard on so many distributions (except Ubuntu, but that's rather a
> > political decision IMHO).
> It does have some good ideas, and it is better than the random bits of
> unmaintained shell it replaces - but it's mediocre at best. No real
> design, just things nailed together with screws and secured with tape.

Which just shows that you probably never seriously dug into
systemd. systemd has a very sensible and mature concept as opposed to
the very hacky System V Init where every distribution has to provide
their *own* init scripts (which clearly shows there is no concept) and
lots of modern functionatity is simply missing.

> > One of the Arch developers actually made a couple of good points why
> > they switched to systemd as default [1].
> >
> Their users really appreciate it, especially those that are now
> migrating to other distros because they preferred their OS when it was
> booting as intended.

Stating from the thread in the Arch forums which I have posted, I
would say that this is simply untrue. People aren't going away from
Arch because of systemd. There are some who are unhappy with it, sure,
but most Arch users support the systemd switch or simply don't care
because they only want their init system to be fast and reliable which
truly is what systemd provides.

Adrian

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121114144...@physik.fu-berlin.de

[Thomas Goirand]

On 11/14/2012 07:53 PM, Matthias Klumpp wrote:
> What are the problems they try to address?

Haven't you read this?
https://lkml.org/lkml/2012/10/2/505

Plus the unwanted move from / to /usr, insane configuration file
things not using /etc, and more RedHat-ismes which have been
discussed at large in this list (let's not restart such thread).

> The strong binding to
> systemd is good and makes much sense to me, and udev is still usable
> without systemd (and will be in the future).

As they merged, it becomes less and less the case. You're seeing
fixes patching udev to rename it systemd-udev, which really, is more
advertizing / marketing than anything, but it shows what kind of
direction its taking. To me, it looks like udev authors are forcing this
so you have no choice but to use systemd.

> Also, both systemd and udev are Linux-only, so the situation here at
> Debian hasn't changed.

Let's say that we choose another implementation of init (let's say,
upstart or OpenRC, or even keep our old sysv-rc), then having
systemd bound to udev and udev bound to systemd will not make
things easy for us.

> The problems we had in the past with bad udev+kernel combinations and
> changing config file format etc. can also be addressed in udev,
> without the need of forking.
> In general, I think a fork of udev would do much more harm than trying
> to solve the problems in udev.

In an ideal world, yes. But when the development goes wild,
and upstream doesn't listen to others or refuses patches,
what kind of alternative do you have?

> Of course, they're free to fork, but
> the separation will hurt both projects and everything relying on
> udev/the fork.

That is correct, but the pain is already there with the new
versions of udev, and its likely it wont change back.

Thomas

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/50A3C10D...@debian.org

[Roger Leigh]

On Wed, Nov 14, 2012 at 03:04:35PM +0100, John Paul Adrian Glaubitz wrote:
> On Wed, Nov 14, 2012 at 09:49:07PM +0800, Patrick Lauer wrote:
> > But anyway, we're getting tired of their ADHD-driven changes just to
> > change things
>
> TBH, I'm getting tired of people who are constantly shooting against
> them because these people are unwilling to accept changes. We're not
> bringing Linux forward if we stick to 30-year-old concepts. systemd is a good
> design and most people actually agree otherwise it wouldn't become
> standard on so many distributions (except Ubuntu, but that's rather a
> political decision IMHO).

systemd does have some good design features. It also has some bad
ones. It's not as black and white as some people have claimed.

If you want a reliable system, you need a reliable PID 1. Putting
additional complexity into PID1 increases the likelihood that a
bug will bring down your *entire system*. PID 1 is a single point
of failure. It *must* be absolutely dependable and reliable.
Upstart is also AFAIK at fault here.

sysvinit is fairly minimal, but even it could be simplified
further. Other init systems (e.g. s6)[1] take that even further
so that at any point in time, PID1 is running an image dedicated
to the current system state, e.g. booting, running, shutting down,
and it will exec() a new image to initiate a state change. When
running normally, PID 1 should do nothing except to reap zombies,
and switch to shutdown. Everything else can be done in a
separate process started by PID 1.

In the case of sysvinit, runlevel changes are delegated to
/etc/init.d/rc. This could be sysv-rc, openrc, or some other
program. If that program fails, init will carry on running.
That's not to say that systemd doesn't do a better job of
resolving dependencies and service management. It does. But.
It's introduced a single point of failure by putting all that
into a single process, running as PID 1. That complexity
should not be in PID1, it should be a separate process. There's
no intrinsic need for it to be there. We've contained the
damage should there be a failure by keeping things separate.

>From a technical POV of how it resolves dependencies and
manages services, systemd should be better. But from the POV of
the system stability and reliability as a whole... that's much
harder to quantify and much less clear cut. After all, if
sysvinit is working for you, and starts up all the services
correctly, once the system is up, it's up. It will continue
to run reliably.


Regards,
Roger

[1] http://www.skarnet.org/software/s6/

--
.''`. Roger Leigh
: :' : Debian GNU/Linux http://people.debian.org/~rleigh/
`. `' schroot and sbuild http://alioth.debian.org/projects/buildd-tools
`- GPG Public Key F33D 281D 470A B443 6756 147C 07B3 C8BC 4083 E800

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/2012111416...@codelibre.net

[John Paul Adrian Glaubitz]

On Nov 14, 2012, at 5:05 PM, Roger Leigh <rle...@codelibre.net> wrote:

> On Wed, Nov 14, 2012 at 03:04:35PM +0100, John Paul Adrian Glaubitz wrote:
>> On Wed, Nov 14, 2012 at 09:49:07PM +0800, Patrick Lauer wrote:
>>> But anyway, we're getting tired of their ADHD-driven changes just to
>>> change things
>>
>> TBH, I'm getting tired of people who are constantly shooting against
>> them because these people are unwilling to accept changes. We're not
>> bringing Linux forward if we stick to 30-year-old concepts. systemd is a good
>> design and most people actually agree otherwise it wouldn't become
>> standard on so many distributions (except Ubuntu, but that's rather a
>> political decision IMHO).
>
> systemd does have some good design features. It also has some bad
> ones. It's not as black and white as some people have claimed.
>
> If you want a reliable system, you need a reliable PID 1. Putting
> additional complexity into PID1 increases the likelihood that a
> bug will bring down your *entire system*. PID 1 is a single point
> of failure. It *must* be absolutely dependable and reliable.
> Upstart is also AFAIK at fault here.

Sticking to the same logic, we should pull out all functionality out of the Linux kernel and use a micro kernel.

Modern computer systems are much more versatile and complex than they were at the time when System V Init was conceived.

You need a certain complexity if you want a certain functionality. I don't want to reboot my computer when changing my network connection, add or remove new hardware like disks or input devices. And I don't want to mess around with configuration files when I want to redirect the audio output of VLC from the internal laptop speakers to an bluetooth or AirPlay.

The reason why Linux has become so successful is because users don't have to mess with tools like isaconf and pnpdump anymore to configure their Soundblaster sound card or edit the interfaces or hosts file to change their IP address.

I honestly think that people who are fighting modern software like systemd, pulse-audio or udev are simply fearing that their expertise in hacking configuration files in order to get things working are no longer needed anymore. They fear that the average joe can install and set up a Linux box without their help.

When I started using Linux in 1998, I would have never thought that I'd be installing it onto my mother's laptop almost 15 years later as the sole operating system and she'd be happily using it with nearly zero support from my side. This would have never been possible without all these little modern helpers that we have nowadays.

If some advanced users want to stick to the traditional Unix way, they're free to use distributions like Gentoo or use any of the BSDs. But I honestly ask them to stop spreading FUD about how software like systemd or Pulse-Audio is hurting Linux and free software, because Linux wouldn't be there where it is nowadays without these developments.

Adrian

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/0277B04F-40FA-4AE1...@physik.fu-berlin.de

[lego...@yandex.ru]

On Wed, Nov 14, 2012 at 06:26:39PM +0100, John Paul Adrian Glaubitz wrote:
> > If you want a reliable system, you need a reliable PID 1. Putting
> > additional complexity into PID1 increases the likelihood that a
> > bug will bring down your *entire system*. PID 1 is a single point
> > of failure. It *must* be absolutely dependable and reliable.
> > Upstart is also AFAIK at fault here.
>
> Sticking to the same logic, we should pull out all functionality out of the Linux kernel and use a micro kernel.
>
> Modern computer systems are much more versatile and complex than they were at the time when System V Init was conceived.

Some things must be as simple as possible even today.

> I honestly think that people who are fighting modern software like systemd, pulse-audio or udev are simply fearing that their expertise in hacking configuration files in order to get things working are no longer needed anymore. They fear that the average joe can install and set up a Linux box without their help.

May be init today should has some new features, but systemd is not such new
init. systemd is a wrong way. See plan9 for a good design examples.

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121114174237.GA31202@localhost

[Thomas Goirand]

On 11/14/2012 10:37 PM, John Paul Adrian Glaubitz wrote:
> True, but as I said, System V Init is not a good concept anymore,
> that's why it's being dropped. Apple dropped the old init system with
> MacOS X 10.4, why should the Linux world still stick to it in 2012?

Could we try not to mix the init system debate with
the udev brokenness one?

Cheers,

Thomas

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/50A3DDB2...@debian.org

[Thomas Goirand]

Hi Adrian,

Roger talks about technical design and upstream author choices,
with some I believe very valid points, and you're talking about
features! This can't go anywhere.

Thomas

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/50A3E07C...@debian.org

[John Paul Adrian Glaubitz]

On Nov 14, 2012, at 7:06 PM, Thomas Goirand <zi...@debian.org> wrote:

> On 11/14/2012 10:37 PM, John Paul Adrian Glaubitz wrote:
>> True, but as I said, System V Init is not a good concept anymore,
>> that's why it's being dropped. Apple dropped the old init system with
>> MacOS X 10.4, why should the Linux world still stick to it in 2012?
> Could we try not to mix the init system debate with
> the udev brokenness one?

udev isn't broken. The only reason why you think udev is broken is because you don't want to use the way it is intended to be used and now you're looking for people to jump the band wagon.

I could start the same discussion regarding other software packages not being compatible with certain platforms: "I want to use $DAEMON on $KERNEL, so please do something, upstream!"

Adrian

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/50FFB4FE-0A85-43F5...@physik.fu-berlin.de

[gustavo panizzo <gfa>]

On 2012-11-14 15:16, John Paul Adrian Glaubitz wrote:
> On Nov 14, 2012, at 7:06 PM, Thomas Goirand <zi...@debian.org> wrote:
>
>> On 11/14/2012 10:37 PM, John Paul Adrian Glaubitz wrote:
>>> True, but as I said, System V Init is not a good concept anymore,
>>> that's why it's being dropped. Apple dropped the old init system
>>> with
>>> MacOS X 10.4, why should the Linux world still stick to it in 2012?
>> Could we try not to mix the init system debate with
>> the udev brokenness one?
>
> udev isn't broken.

really?

https://bbs.archlinux.org/viewtopic.php?id=134012&p=1

but don't trust me

https://lkml.org/lkml/2012/10/2/505

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/973bc827353d2f58...@zumbi.com.ar

[Guillem Jover]

Hi!

On Wed, 2012-11-14 at 21:49:07 +0800, Patrick Lauer wrote:
> But anyway, we're getting tired of their ADHD-driven changes just to
> change things - maybe we can build up enough momentum so that things
> might just be less frustrating for us all. You're all welcome to join,
> ignore us or do what you want.

Nice! I've had on my TODO (but TBH as a very low priority item), to
look into something like this too, but I was thinking about taking
FreeBSD's devd and trying to port it to other systems instead.

In any case I'll keep an eye on this fork, and I'll gladly switch to
it whenever it's available.

You might want to contact the Ubuntu folks too, as I don't think they
are planning on switching to systemd any time soon.

thanks,
guillem

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121114193...@gaara.hadrons.org

[Steve Langasek]

On Wed, Nov 14, 2012 at 03:04:35PM +0100, John Paul Adrian Glaubitz wrote:
> On Wed, Nov 14, 2012 at 09:49:07PM +0800, Patrick Lauer wrote:
> > But anyway, we're getting tired of their ADHD-driven changes just to
> > change things

> TBH, I'm getting tired of people who are constantly shooting against
> them because these people are unwilling to accept changes. We're not
> bringing Linux forward if we stick to 30-year-old concepts. systemd is a
> good design and most people actually agree otherwise it wouldn't become
> standard on so many distributions (except Ubuntu, but that's rather a
> political decision IMHO).

Pretty sure you have this backwards. The decision to implement upstart and
use it in Ubuntu was a political one. The decision to NIH a
dependency-based init system and then try to strongarm everyone into using
it by breaking compatibility was the political one.

BTW, if systemd is a good design, why does it rely so heavily on
socket-based activation, which has fundamentally unmaintainable security?

> One of the Arch developers actually made a couple of good points why
> they switched to systemd as default [1].

I have spent too much time arguing against the perceived deficiencies of
systemd (such as [...] "it was started by Lennart Poettering" [...]). I
strongly believe that 1) all of these perceived deficiencies are not
deficiencies, but are actually benefits [...]

So I guess the Arch developers also have nothing but nice things to say
about pulseaudio and avahi.

--
Steve Langasek Give me a lever long enough and a Free OS
Debian Developer to set it on, and I can move the world.
Ubuntu Developer http://www.debian.org/
slan...@ubuntu.com vor...@debian.org

[Steve Langasek]

On Wed, Nov 14, 2012 at 12:47:11PM -0800, Steve Langasek wrote:
> On Wed, Nov 14, 2012 at 03:04:35PM +0100, John Paul Adrian Glaubitz wrote:
> > On Wed, Nov 14, 2012 at 09:49:07PM +0800, Patrick Lauer wrote:
> > > But anyway, we're getting tired of their ADHD-driven changes just to
> > > change things

> > TBH, I'm getting tired of people who are constantly shooting against
> > them because these people are unwilling to accept changes. We're not
> > bringing Linux forward if we stick to 30-year-old concepts. systemd is a
> > good design and most people actually agree otherwise it wouldn't become
> > standard on so many distributions (except Ubuntu, but that's rather a
> > political decision IMHO).

> Pretty sure you have this backwards. The decision to implement upstart and
> use it in Ubuntu was a political one.

Haha, I mean technical. ;)

[John Paul Adrian Glaubitz]

On Nov 14, 2012, at 10:11 PM, Steve Langasek <vor...@debian.org> wrote:

> On Wed, Nov 14, 2012 at 12:47:11PM -0800, Steve Langasek wrote:
>> On Wed, Nov 14, 2012 at 03:04:35PM +0100, John Paul Adrian Glaubitz wrote:
>>> On Wed, Nov 14, 2012 at 09:49:07PM +0800, Patrick Lauer wrote:
>>>> But anyway, we're getting tired of their ADHD-driven changes just to
>>>> change things
>
>>> TBH, I'm getting tired of people who are constantly shooting against
>>> them because these people are unwilling to accept changes. We're not
>>> bringing Linux forward if we stick to 30-year-old concepts. systemd is a
>>> good design and most people actually agree otherwise it wouldn't become
>>> standard on so many distributions (except Ubuntu, but that's rather a
>>> political decision IMHO).
>
>> Pretty sure you have this backwards. The decision to implement upstart and
>> use it in Ubuntu was a political one.
>
> Haha, I mean technical. ;)

Haha, gotcha! :)

Adrian

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/39499AC6-5C31-4541...@physik.fu-berlin.de

[Uoti Urpala]

Steve Langasek wrote:
> Pretty sure you have this backwards. The decision to implement upstart and

> use it in Ubuntu was a technical [corrected] one. The decision to NIH a

> dependency-based init system and then try to strongarm everyone into using
> it by breaking compatibility was the political one.

The decision to create upstart was a technical decision. However,
upstart had design flaws, and so systemd was created to do better. This
was also a technical decision. Do you seriously claim that it would have
been possible to work within the existing upstart project to bring it to
the level of current systemd? I find that totally implausible.

Ubuntu still sticking to upstart is a political decision as far as I can
see; there is no technical reason why it would be a better alternative
even for their own use than systemd.

> BTW, if systemd is a good design, why does it rely so heavily on
> socket-based activation, which has fundamentally unmaintainable security?

What exactly do you mean by this "fundamentally unmaintainable security"
claim?

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/1352929546.1...@glyph.nonexistent.invalid

[Matthias Klumpp]

Hi!
All these concerns about systemd and systemd vs. upstart have been
addressed in a very good way by the systemd authors.
Also, I would like to point out that "systemd" is the name of a
project with multiple binaries - all the features systemd provides
don't mean that everything is running in one process, in fact, systemd
spawns some helpers on-demand for many actions. (However, the systemd
core does more things than SysVInit, but as long as this code is
tested and works well (and isn't changed that often) I have no
concern)
Also, systemd provides systemd-logind, an excellent way to get rid of
ConsoleKit, which also makes it possible to have real multiseat
support. And managing services using systemd is fantastic! :)

But well, back to udev: I am not personally involved with the udev
packaging, but has someone already talked to the people making the
criticised decisions to explain themselves? I don't think ReadHat
developers want to do any damage to these integral components, so I
hardly think that there were no reasons for a change.
Also, if these issues cannot be solved, would maintaining a small
patchset for udev be an option?
Cheers,
Matthias

2012/11/14 Uoti Urpala <uoti....@pp1.inet.fi>:

Archive: http://lists.debian.org/CAKNHny9eocGg7xjYxo_duhihox=r60b_Uf7Tpw...@mail.gmail.com

[John Paul Adrian Glaubitz]

On Wed, Nov 14, 2012 at 03:41:55PM -0300, gustavo panizzo <gfa> wrote:
> >udev isn't broken.
>
> really?
>
> https://bbs.archlinux.org/viewtopic.php?id=134012&p=1

I actually remember having seen this issue on Fedora Rawhide as well,
but it vanished after an update a few weeks ago, so it rather seems
like a "normal" bug to me. That's not really what "broken" means in
this context.

> but don't trust me
>
> https://lkml.org/lkml/2012/10/2/505

Well, yes, it's the same issue. Linus is well known for going on a
rant very quickly, but that doesn't mean that udev is completely
broken.

Yes, they obviously made a recent change that broke module loading on
some machines, but that doesn't mean the whole concept is
broken. That's just an unfair statement. Also, Kay is admitting that
there is/was a problem with udev that needs to be addressed and it
seems that they did because I cannot reproduce it anymore with udev
195 anymore.

Adrian

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121114223...@physik.fu-berlin.de

[Ben Hutchings]

On Wed, Nov 14, 2012 at 11:34:27PM +0100, John Paul Adrian Glaubitz wrote:
> On Wed, Nov 14, 2012 at 03:41:55PM -0300, gustavo panizzo <gfa> wrote:
> > >udev isn't broken.
> >
> > really?
> >
> > https://bbs.archlinux.org/viewtopic.php?id=134012&p=1
>
> I actually remember having seen this issue on Fedora Rawhide as well,
> but it vanished after an update a few weeks ago, so it rather seems
> like a "normal" bug to me. That's not really what "broken" means in
> this context.
>
> > but don't trust me
> >
> > https://lkml.org/lkml/2012/10/2/505
>
> Well, yes, it's the same issue. Linus is well known for going on a
> rant very quickly, but that doesn't mean that udev is completely
> broken.
>
> Yes, they obviously made a recent change that broke module loading on
> some machines, but that doesn't mean the whole concept is
> broken. That's just an unfair statement. Also, Kay is admitting that
> there is/was a problem with udev that needs to be addressed and it
> seems that they did because I cannot reproduce it anymore with udev
> 195 anymore.

I believe the regression (removal of support for firmware loading
during module loading) has been fixed. However, the udev developers
*knew in advance* that this would be a problem, reported such uses
of firmware loading as being driver bugs. They then went ahead and
changed udev even though the drivers had not all been updated (and it
was evidently not easy to do so in some cases).

Ben.

--
Ben Hutchings
We get into the habit of living before acquiring the habit of thinking.
- Albert Camus

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121114224...@decadent.org.uk

[Pau Garcia i Quiles]

On Wed, Nov 14, 2012 at 11:43 PM, Ben Hutchings <b...@decadent.org.uk> wrote:

> I believe the regression (removal of support for firmware loading
> during module loading) has been fixed. However, the udev developers
> *knew in advance* that this would be a problem, reported such uses
> of firmware loading as being driver bugs. They then went ahead and
> changed udev even though the drivers had not all been updated (and it
> was evidently not easy to do so in some cases).

If the systemd/udev people continue with that attitude, I'm expecting
Linus Torvalds to come with his super-simple, super-fast,
super-minimal and super-brilliant udev replacement any minute now.

--
Pau Garcia i Quiles
http://www.elpauer.org
(Due to my workload, I may need 10 days to answer)

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/CAKcBokuor9JEyhVh0-4bn=6xoh693vcvy4wf...@mail.gmail.com

[Steve Langasek]

On Wed, Nov 14, 2012 at 04:05:12PM +0000, Roger Leigh wrote:
> On Wed, Nov 14, 2012 at 03:04:35PM +0100, John Paul Adrian Glaubitz wrote:
> > On Wed, Nov 14, 2012 at 09:49:07PM +0800, Patrick Lauer wrote:
> > > But anyway, we're getting tired of their ADHD-driven changes just to
> > > change things

> > TBH, I'm getting tired of people who are constantly shooting against
> > them because these people are unwilling to accept changes. We're not
> > bringing Linux forward if we stick to 30-year-old concepts. systemd is a good
> > design and most people actually agree otherwise it wouldn't become
> > standard on so many distributions (except Ubuntu, but that's rather a
> > political decision IMHO).

> systemd does have some good design features. It also has some bad
> ones. It's not as black and white as some people have claimed.

> If you want a reliable system, you need a reliable PID 1. Putting
> additional complexity into PID1 increases the likelihood that a
> bug will bring down your *entire system*. PID 1 is a single point
> of failure. It *must* be absolutely dependable and reliable.
> Upstart is also AFAIK at fault here.

[Citation needed]

Upstart provides a PID 1 that is absolutely rock solid. It's true that it's
more complex than sysvinit, because it's more featureful; but great care has
been taken to only pull the features into PID 1 that absolutely have to be
there, and the implementation of those features is very elegant and
maintainable.[1]

Aside from libc, upstart has only two external library dependencies (three
in trunk), dbus and nih:

$ objdump -p /sbin/init | grep NEEDED
NEEDED libnih.so.1
NEEDED libnih-dbus.so.1
NEEDED libdbus-1.so.3
NEEDED librt.so.1
NEEDED libc.so.6
$

And upstart is rigorously unit-tested at build time.

That's a far cry from systemd's 8 external dependencies:

$ objdump -p /lib/systemd/systemd | grep NEEDED
NEEDED libselinux.so.1
NEEDED libdbus-1.so.3
NEEDED libudev.so.0
NEEDED libwrap.so.0
NEEDED libpam.so.0
NEEDED libaudit.so.0
NEEDED libcap.so.2
NEEDED libkmod.so.2
NEEDED librt.so.1
NEEDED libc.so.6
NEEDED ld-linux.so.2
$

And of all the concerns raised when Ubuntu (and Fedora and OpenSuSE)
switched to upstart, "PID 1 is buggy and crashes" was not one of them.

> sysvinit is fairly minimal, but even it could be simplified
> further. Other init systems (e.g. s6)[1] take that even further
> so that at any point in time, PID1 is running an image dedicated
> to the current system state, e.g. booting, running, shutting down,
> and it will exec() a new image to initiate a state change. When
> running normally, PID 1 should do nothing except to reap zombies,
> and switch to shutdown. Everything else can be done in a
> separate process started by PID 1.

This is an arbitrary design constraint that's not grounded in anyone's
actual experience of deploying upstart.

This is not theoretical. upstart has been PID 1 in Ubuntu since 2006. It
*is* absolutely dependable and reliable.

--
Steve Langasek Give me a lever long enough and a Free OS
Debian Developer to set it on, and I can move the world.
Ubuntu Developer http://www.debian.org/
slan...@ubuntu.com vor...@debian.org

[1] http://ifdeflinux.blogspot.co.uk/2012/11/upstart-cookbook-updated-for-developers.html

[John Paul Adrian Glaubitz]

On Nov 15, 2012, at 12:23 AM, Steve Langasek <vor...@debian.org> wrote:

Upstart provides a PID 1 that is absolutely rock solid.  It's true that it's
more complex than sysvinit, because it's more featureful;

The same is valid for the comparision of upstart vs systemd.

And of all the concerns raised when Ubuntu (and Fedora and OpenSuSE)
switched to upstart, "PID 1 is buggy and crashes" was not one of them.

With only Ubuntu being the remaining distribution sticking to upstart, while nearly everyone else has switched to systemd.

This is not theoretical.  upstart has been PID 1 in Ubuntu since 2006.  It
*is* absolutely dependable and reliable.

Upstart has had its problems, too [1]. And, honestly, the way this bug was handled left me with little confidence in upstart.

Adrian

> [1] https://bugs.launchpad.net/ubuntu/+source/upstart/+bug/557177

[John Paul Adrian Glaubitz]

On Nov 14, 2012, at 6:43 PM, lego...@yandex.ru wrote:

>> Modern computer systems are much more versatile and complex than they were at the time when System V Init was conceived.
>
> Some things must be as simple as possible even today.

Care to elaborate why? To save memory on an 8 GB workstation? Even the 25 US$ Raspberry Pi has enough power for systemd.

Are you also choosing FAT32 over ext4 because it is simpler?

Why are we having Fibre Channel support in the kernel? Why does the kernel include a virtual machine hypervisor? Why do we support IPv6?

We could just go back and stick with our good old SunOS 4 boxes.

>
>> I honestly think that people who are fighting modern software like systemd, pulse-audio or udev are simply fearing that their expertise in hacking configuration files in order to get things working are no longer needed anymore. They fear that the average joe can install and set up a Linux box without their help.
>
> May be init today should has some new features, but systemd is not such new
> init. systemd is a wrong way. See plan9 for a good design examples.

What makes you think that systemd does it the wrong way? They are using a very similar concept that Apple uses very successfully on MacOS X since 10.4 while no one in this universe has ever touched Plan 9 again.

People are constantly insisting that systemd is too bloated or unreliable, but yet no one has really come up with real examples to prove that.

Yes, the core binary of System V Init is smaller than systemd's. However, System V Init needs a lot of bloat in form of hacky bash scripts using even more external tools like sed and awk to be actually useful in any regard.

And I think it makes way more sense to have all the functionality of the init system integrated into it's core binary rather than depending on external scripts which will hopefully do what init expects from them.

Adrian

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/290B451F-CC15-4793...@physik.fu-berlin.de

[Ben Hutchings]

On Thu, Nov 15, 2012 at 12:45:48AM +0100, John Paul Adrian Glaubitz wrote:
> On Nov 15, 2012, at 12:23 AM, Steve Langasek <vor...@debian.org> wrote:

[...]

> > This is not theoretical. upstart has been PID 1 in Ubuntu since 2006. It
> > *is* absolutely dependable and reliable.
>
> Upstart has had its problems, too [1]. And, honestly, the way this bug was handled left me with little confidence in upstart.
>
> Adrian
>
> > [1] https://bugs.launchpad.net/ubuntu/+source/upstart/+bug/557177

I suppose you should comment on it too, just to add your indignation
at a bug that never affected you and wasn't fixed for a whole 2 days.

Ben.

--
Ben Hutchings
We get into the habit of living before acquiring the habit of thinking.
- Albert Camus

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121115001...@decadent.org.uk

[Steve Langasek]

On Thu, Nov 15, 2012 at 12:45:48AM +0100, John Paul Adrian Glaubitz wrote:
> > This is not theoretical. upstart has been PID 1 in Ubuntu since 2006. It
> > *is* absolutely dependable and reliable.

> Upstart has had its problems, too [1].

> [1] https://bugs.launchpad.net/ubuntu/+source/upstart/+bug/557177

A bug in an upstart job, and your dissatisfaction with how the upstart
maintainer responded to the bug report, is entirely orthogonal to Roger's
point about complexity in PID 1.

> And, honestly, the way this bug was handled left me with little confidence
> in upstart.

I'm very sad for you.

[John Paul Adrian Glaubitz]

On Nov 15, 2012, at 1:17 AM, Ben Hutchings <b...@decadent.org.uk> wrote:

>>> [1] https://bugs.launchpad.net/ubuntu/+source/upstart/+bug/557177
>
> I suppose you should comment on it too, just to add your indignation
> at a bug that never affected you and wasn't fixed for a whole 2 days.

Yes, it was fixed, after a very heated discussion with the maintainer who was blaiming users first for using the script incorrectly.

Again, that's what can happen if you rely on hacky bash scripts for core functionality.

Adrian

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/77EE56AA-6C1C-4CDA...@physik.fu-berlin.de

[John Paul Adrian Glaubitz]

On Nov 15, 2012, at 1:19 AM, Steve Langasek <vor...@debian.org> wrote:

> On Thu, Nov 15, 2012 at 12:45:48AM +0100, John Paul Adrian Glaubitz wrote:
>>> This is not theoretical. upstart has been PID 1 in Ubuntu since 2006. It
>>> *is* absolutely dependable and reliable.
>
>> Upstart has had its problems, too [1].
>
>> [1] https://bugs.launchpad.net/ubuntu/+source/upstart/+bug/557177
>
> A bug in an upstart job, and your dissatisfaction with how the upstart
> maintainer responded to the bug report, is entirely orthogonal to Roger's
> point about complexity in PID 1.

I don't think so. It shows what can happen if you delagate fundamental tasks out of the core init binary into external bash scripts.

Upstart has to rely entirely rely on the external script to do the right thing instead of doing it itself. You are constantly argueing that this makes the whole system more reliable, yet it took one apparently harmless command to kill the entire filesystem.

I don't want to imagine this situation on our backup or home directory server.

This would not have happened with systemd's design.

>
>> And, honestly, the way this bug was handled left me with little confidence
>> in upstart.
>
> I'm very sad for you.

So, you think marking such a major flaw as a wishlist is an appropriate reaction?

Adrian

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/1A08D899-27F3-474F...@physik.fu-berlin.de

[Matthew Thode]

As a source for some of our concerns here is us trying to separate out
the udev build system so we can build ONLY udev if we want to install
ONLY udev (we have to build systemd if we want ONLY udev right now).

This means we have to pull in build deps even if we don't actually need
them.

http://lists.freedesktop.org/archives/systemd-devel/2012-June/005464.html


--
-- Matthew Thode (prometheanfire)

[Roger Leigh]

On Thu, Nov 15, 2012 at 12:57:50AM +0100, John Paul Adrian Glaubitz wrote:
> On Nov 14, 2012, at 6:43 PM, lego...@yandex.ru wrote:
>
> >> Modern computer systems are much more versatile and complex than they were at the time when System V Init was conceived.
> >
> > Some things must be as simple as possible even today.
>
> Care to elaborate why? To save memory on an 8 GB workstation? Even the 25 US$ Raspberry Pi has enough power for systemd.

It's very simple. What happens if the init process terminates?

The answer is that you get an instant kernel panic. PID 1 must
never die.

Test it yourself: boot with 'init=/bin/bash' and then type 'exit'
to terminate the shell.

So if the init process receives a signal like a SIGSEGV due to
tripping over a bad pointer, your system will die immediately.
Therefore, keeping PID 1 as simple as reasonably possible is of
critical importance. [OK, you can try to mitigate by re-execing
yourself in a signal handler, but even that adds extra complexity
and is itself not without danger. I hope you take the point.]

So systems which put additional logic in PID 1 are going to increase
the probability of bugs being present, and those bugs could kill
your system. There is no need for systemd, upstart, or any init
system to have anything more than the bare minimum in PID 1; you
can just fork and exec the more complicated part and keep this
separated.

So it's nothing to do about how powerful the system is. Or even
if we're running unit files, upstart jobs or shell scripts. It's
to do with the fundamental reliability of PID 1, because this is
a critical point of failure; if it dies, there's no recovery, the
system is dead. If you had to run a system which was safety
critical, you wouldn't run systemd on it, and you wouldn't run
upstart. Even if they were tested extensively, it's just too great
a risk. If you were really serious, you'd probably not run
sysvinit either; it's better in this respect than the other two, but
there are still tinier, more easily verifiable init systems out there
where it's just a screenful of code, and it's provably correct.


Regards,
Roger

--
.''`. Roger Leigh
: :' : Debian GNU/Linux http://people.debian.org/~rleigh/
`. `' schroot and sbuild http://alioth.debian.org/projects/buildd-tools
`- GPG Public Key F33D 281D 470A B443 6756 147C 07B3 C8BC 4083 E800

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/2012111501...@codelibre.net

[Ben Hutchings]

On Thu, Nov 15, 2012 at 01:11:32AM +0000, Roger Leigh wrote:
[...]

> So systems which put additional logic in PID 1 are going to increase
> the probability of bugs being present, and those bugs could kill
> your system.

[...]

This is also true for the kernel, which is why we generally prefer
to use Hurd... or not.

Ben.

--
Ben Hutchings
We get into the habit of living before acquiring the habit of thinking.
- Albert Camus

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121115020...@decadent.org.uk

[Thomas Goirand]

On 11/15/2012 06:43 AM, Ben Hutchings wrote:
> I believe the regression (removal of support for firmware loading
> during module loading) has been fixed. However, the udev developers
> *knew in advance* that this would be a problem, reported such uses of
> firmware loading as being driver bugs. They then went ahead and
> changed udev even though the drivers had not all been updated (and it
> was evidently not easy to do so in some cases). Ben.

Which is the exact same attitude as for moving to /usr,
refusing patches for supporting other arch, and so on.
They just don't care about breaking other people's
system. This *will* happen again.

Thomas

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/50A460A6...@debian.org

[Thomas Goirand]

On 11/15/2012 10:07 AM, Ben Hutchings wrote:
> On Thu, Nov 15, 2012 at 01:11:32AM +0000, Roger Leigh wrote:
> [...]
>> So systems which put additional logic in PID 1 are going to increase
>> the probability of bugs being present, and those bugs could kill
>> your system.
> [...]
>
> This is also true for the kernel, which is why we generally prefer
> to use Hurd... or not.
>
> Ben.

The fact we aren't using Hurd has *nothing* to do with
the fact it is a micro kernel, and you know it.

If Hurd has the same level of hardware support as
Linux, as many contributors, and as many features,
then probably, it would also have as many users.

Mixing the discussion around feature and engineer
design does *not* work, even in the case of kernels.

Thomas

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/50A46377...@debian.org

[Thomas Goirand]

On 11/15/2012 07:57 AM, John Paul Adrian Glaubitz wrote:
> On Nov 14, 2012, at 6:43 PM, lego...@yandex.ru wrote:
>
>>> Modern computer systems are much more versatile and complex than they were at the time when System V Init was conceived.
>> Some things must be as simple as possible even today.
> Care to elaborate why? To save memory on an 8 GB workstation? Even the 25 US$ Raspberry Pi has enough power for systemd.
>
> Are you also choosing FAT32 over ext4 because it is simpler?
>
> Why are we having Fibre Channel support in the kernel? Why does the kernel include a virtual machine hypervisor? Why do we support IPv6?
>
> We could just go back and stick with our good old SunOS 4 boxes.

Would you care reading what's being written to you?
Roger didn't write that the init system should be kept
without features, using old age techniques. He wrote
that he believes that *pid 1* should be kept simple,
and that the complexity should go somewhere else
(like, in a fork of PID 1, for example). That is very
different from what you write above.

>> People are constantly insisting that systemd is too bloated or unreliable, but yet no one has really come up with real examples to prove that.

I'm sorry but the sentence above is just plain wrong.
See Linus post, and all the things that udev broke.

>> Yes, the core binary of System V Init is smaller than systemd's. However, System V Init needs a lot of bloat in form of hacky bash scripts using even more external tools like sed and awk to be actually useful in any regard.

But these hacks / bloats will not ultimately result in a
kernel crash.

>> And I think it makes way more sense to have all the functionality of the init system

Yes.

>> integrated into it's core binary

And no! :)

It would have been possible to do the right thing (tm)
and have both feature and reliability. Currently, we
only have the former, which is due to the design.

Thomas

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/50A4654D...@debian.org

[Ben Hutchings]

On Thu, 2012-11-15 at 11:37 +0800, Thomas Goirand wrote:
> On 11/15/2012 10:07 AM, Ben Hutchings wrote:
> > On Thu, Nov 15, 2012 at 01:11:32AM +0000, Roger Leigh wrote:
> > [...]
> >> So systems which put additional logic in PID 1 are going to increase
> >> the probability of bugs being present, and those bugs could kill
> >> your system.
> > [...]
> >
> > This is also true for the kernel, which is why we generally prefer
> > to use Hurd... or not.
> >
> > Ben.
>
> The fact we aren't using Hurd has *nothing* to do with
> the fact it is a micro kernel, and you know it.

Aside from the fact that micro-kernels are grossly impractical.

> If Hurd has the same level of hardware support as
> Linux, as many contributors, and as many features,
> then probably, it would also have as many users.

It turns out that stupid architectures make it hard to attract and keep
contributors.

> Mixing the discussion around feature and engineer
> design does *not* work, even in the case of kernels.

Whether the kernel or init survives a crash is completely unimportant if
the applications the computer is supposed to run become unavailable.
What good is a smaller, less buggy init if it can't keep (say) apache
and sshd running?

Ben.

--
Ben Hutchings
friends: People who know you well, but like you anyway.

[Oleg]

On Thu, Nov 15, 2012 at 12:57:50AM +0100, John Paul Adrian Glaubitz wrote:
> > Some things must be as simple as possible even today.
>
> Care to elaborate why? To save memory on an 8 GB workstation? Even the 25 US$ Raspberry Pi has enough power for systemd.

This is obvious. For security and stability reasons. This is KISS.

> Are you also choosing FAT32 over ext4 because it is simpler?

Yes, of course. In some embedded devices we use fat32 or ext2. It simpler
and faster.

> > May be init today should has some new features, but systemd is not such new
> > init. systemd is a wrong way. See plan9 for a good design examples.
>
> What makes you think that systemd does it the wrong way? They are using a very similar concept that Apple uses very successfully on MacOS X since 10.4 while no one in this universe has ever touched Plan 9 again.

Who said that Apple concepts are technically good? I don't think so.

> People are constantly insisting that systemd is too bloated or unreliable, but yet no one has really come up with real examples to prove that.

I think this is the question of the near future.

> Yes, the core binary of System V Init is smaller than systemd's. However, System V Init needs a lot of bloat in form of hacky bash scripts using even more external tools like sed and awk to be actually useful in any regard.

And what? The easy and power extension mechanisms are bad? I don't
understand, why do people that don't like and don't understand unix ideas
still use it and complain about it?

> And I think it makes way more sense to have all the functionality of the init system integrated into it's core binary rather than depending on external scripts which will hopefully do what init expects from them.

Sorry, but this is not true. This is the bad design and a wrong way.

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121115061159.GA7358@localhost

[Wouter Verhelst]

On Wed, Nov 14, 2012 at 03:37:21PM +0100, John Paul Adrian Glaubitz wrote:
> On Wed, Nov 14, 2012 at 09:20:05AM -0500, The Wanderer wrote:
> > But why is a 30-year-old concept necessarily worse than a new one? Or to put it
> > another way, why is it necessary to "bring Linux forward", in cases where what
> > is already present is good and works well? (And, taken further: in cases where
> > what is already there *isn't* good and/or *doesn't* work well, why is it
> > necessary to accept change *in a particular direction*, if that direction has
> > problems of its own?)
>
> Because System V Init isn't a good concept.

This I disagree with.

It has problems, yes. But it's been able to boot billions of machines
for several decades. That can't mean it's not good. Things that aren't
good can't manage to do such things.

There are problems with systemd too, and they're not limited to "the
developers are obnoxious".

--
Copyshops should do vouchers. So that next time some bureaucracy requires you
to mail a form in triplicate, you can mail it just once, add a voucher, and
save on postage.

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121115081...@grep.be

[Guillem Jover]

On Wed, 2012-11-14 at 15:23:51 -0800, Steve Langasek wrote:
> On Wed, Nov 14, 2012 at 04:05:12PM +0000, Roger Leigh wrote:
> > If you want a reliable system, you need a reliable PID 1. Putting
> > additional complexity into PID1 increases the likelihood that a
> > bug will bring down your *entire system*. PID 1 is a single point
> > of failure. It *must* be absolutely dependable and reliable.
> > Upstart is also AFAIK at fault here.
>
> [Citation needed]
>
> Upstart provides a PID 1 that is absolutely rock solid. It's true that it's
> more complex than sysvinit, because it's more featureful; but great care has
> been taken to only pull the features into PID 1 that absolutely have to be
> there, and the implementation of those features is very elegant and
> maintainable.[1]
>
> Aside from libc, upstart has only two external library dependencies (three
> in trunk), dbus and nih:
>
> $ objdump -p /sbin/init | grep NEEDED

...
> NEEDED libnih-dbus.so.1
> NEEDED libdbus-1.so.3
...

> $
>
> And upstart is rigorously unit-tested at build time.
>
> That's a far cry from systemd's 8 external dependencies:
>
> $ objdump -p /lib/systemd/systemd | grep NEEDED

...
> NEEDED libdbus-1.so.3
...
> $

TBH, I'd not trust my system to *any* critical service that uses dbus,
AFAIK it still asserts on error conditions (including non-programmer
errors). Whenever I've had to code a critical service that needed to
use dbus, I've confined its execution to a subprocess.

regards,
guillem

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/2012111508...@gaara.hadrons.org

[Chow Loong Jin]

On 15/11/2012 16:14, Wouter Verhelst wrote:
> It has problems, yes. But it's been able to boot billions of machines
> for several decades. That can't mean it's not good. Things that aren't
> good can't manage to do such things.

Oh sure, it isn't bad, it just isn't good enough. I don't think machines several
decades back booted particularly quickly, supported hotplug, or had any of the
bells and whistles we have today. While we're at it let's go back and use SCO
Unix, shall we?

--
Kind regards,
Loong Jin

[Tollef Fog Heen]

]] Roger Leigh

> So if the init process receives a signal like a SIGSEGV due to
> tripping over a bad pointer, your system will die immediately.

No, it does not. With init, you can catch it and continue. In the case
of systemd, it dumps core (if possible) and then freezes itself so it
stops doing anything such as reaping zombies and you need to reboot
quite quickly, but the system continues to mostly work.

--
Tollef Fog Heen
UNIX is user friendly, it's just picky about who its friends are

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/87zk2jb...@qurzaw.varnish-software.com

[Chow Loong Jin]

AFAIK g_assert() and assert() get compiled into nothing when debugging is turned
off, which is what's done for release builds.

[Guillem Jover]

dbus uses its own assert function (_dbus_real_assert(), not based on
either assert() or g_assert()), when DBUS_DISABLE_ASSERT is undefined.

In any case, you are right, its use seems to be only enabled on debug
builds in Debian (--enable-asserts used on debian/rules), which is
good, AFAIR this did not use to be the case in the past.

thanks,

guillem


--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121115092...@gaara.hadrons.org

[Nikolaus Rath]

Hmm. Without further information, that could just as well be interpreted
as upstart containing 5 less tested reimplementations of existing
libraries.



Best,

-Nikolaus

--
»Time flies like an arrow, fruit flies like a Banana.«

PGP fingerprint: 5B93 61F8 4EA2 E279 ABF6 02CF A9AD B7F8 AE4E 425C

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/87txsqg...@inspiron.ap.columbia.edu

[Thomas Goirand]

On 11/14/2012 11:12 AM, Thomas Goirand wrote:
> Hi,
>
> I think this is an interesting read:
> http://article.gmane.org/gmane.linux.gentoo.project/2262
>
> The full thread is here:
> http://thread.gmane.org/gmane.linux.gentoo.project/2262
>
> As Gentoo guys and some major kernel people are protesting about the
> insanity Kay and Lennart have done to udev, probably this is a window of
> opportunity to stop being annoyed in Debian, by brutal RedHat decisions
> which we don't necessarily (have to) agree on (eg: move to /usr, merge
> with systemd, etc.).
>
> Thoughts anyone? Would anyone have time to work with Gentoo? What is the
> view of the kernel team and the udev (marco?) maintainers?
>
> Cheers,
>
> Thomas
This thread was originally about udev, yet everyone is starting again the
systemd / upstart / sysv-rc war. I think we can agree that we don't about
the init system, and it wasn't my intention to restart this debate.

However, how should Debian see this udev fork? Could it be beneficial to
Debian somehow? Is there the will to maintain it? Are the udev (or kernel)
maintainers opposing this (future) fork to enter Debian? If yes, on what
ground? Is it even realistic to think Debian could use such a fork, which
probably the upstream kernel maintainers wont support? What is the view
of kernel.org people on this btw?

I still have no answer to these... Yet udev is a crucial piece of software,
and I've been told in this list that mdev is a toy that isn't fit for
the job.

Thomas

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/50A52FEF...@debian.org

[Russ Allbery]

Thomas Goirand <zi...@debian.org> writes:

> This thread was originally about udev, yet everyone is starting again
> the systemd / upstart / sysv-rc war. I think we can agree that we don't
> about the init system, and it wasn't my intention to restart this
> debate.

> However, how should Debian see this udev fork? Could it be beneficial to
> Debian somehow? Is there the will to maintain it? Are the udev (or
> kernel) maintainers opposing this (future) fork to enter Debian? If yes,
> on what ground? Is it even realistic to think Debian could use such a
> fork, which probably the upstream kernel maintainers wont support? What
> is the view of kernel.org people on this btw?

I think part of the reason why the conversation hasn't been very
productive is that this is somewhat premature. At this point, I think the
only reason to know about the fork (still a significant reason, don't get
me wrong!) is so that anyone in Debian who feels motivated to help out
with the fork can join the upstream development team.

In terms of it being a viable choice for integration into Debian, it's way
too early to know one way or the other, and there's really no point in
speculating (which is why the speculation goes off into unproductive
corners like the long-standing init system argument). There are things
that could happen that would make it clearly a bad choice (the fork
doesn't stay viable enough to produce releases) or clearly a good choice
(most of the existing udev developers end up contributing to the fork and
then it becomes the official udev, similar to what happened to gcc in the
egcs days). The debate is mostly over the likelihood of those and other
intermediate possibilities, and rather than arguing about it, we could
just wait and see. :)

--
Russ Allbery (r...@debian.org) <http://www.eyrie.org/~eagle/>

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/87y5i23...@windlord.stanford.edu

[Kelly Clowers]

On Wed, Nov 14, 2012 at 12:47 PM, Steve Langasek <vor...@debian.org> wrote:
>
> BTW, if systemd is a good design, why does it rely so heavily on
> socket-based activation, which has fundamentally unmaintainable security?

Please excuse the intrusion of a mere user, but I haven't heard that
one yet, can you elaborate?

Cheers,
Kelly Clowers

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/CAFoWM=_0YhxEEpp6CgptZKzEeC=8w=2=+sx8CHni8...@mail.gmail.com

[Salvo Tomaselli]

On Thursday 15 November 2012 00.57.50 John Paul Adrian Glaubitz wrote:
> People are constantly insisting that systemd is too bloated or unreliable,
> but yet no one has really come up with real examples to prove that.

Hum, actually when i tried it, i couldn't halt or reboot my machine without an
hard reset. I suppose that is a real example of bloated?

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=661239

Bye
--
Salvo Tomaselli

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/201211162345....@tiscali.it

[John Paul Adrian Glaubitz]

On Fri, Nov 16, 2012 at 11:45:45PM +0100, Salvo Tomaselli wrote:
> On Thursday 15 November 2012 00.57.50 John Paul Adrian Glaubitz wrote:
> > People are constantly insisting that systemd is too bloated or unreliable,
> > but yet no one has really come up with real examples to prove that.
>
> Hum, actually when i tried it, i couldn't halt or reboot my machine without an
> hard reset. I suppose that is a real example of bloated?
>
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=661239

Hmm, that doesn't look like a valid bug report to me. Especially I
don't see why dhclient would be able to disrupt systemd in such a way
that you'd need to do a hard reboot.

Adrian

--
.''`. John Paul Adrian Glaubitz
: :' : Debian Developer - glau...@debian.org
`. `' Freie Universitaet Berlin - glau...@physik.fu-berlin.de
`- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121116230...@physik.fu-berlin.de

[Kelly Clowers]

On Fri, Nov 16, 2012 at 2:45 PM, Salvo Tomaselli <tipo...@tiscali.it> wrote:
> On Thursday 15 November 2012 00.57.50 John Paul Adrian Glaubitz wrote:
>> People are constantly insisting that systemd is too bloated or unreliable,
>> but yet no one has really come up with real examples to prove that.
>
> Hum, actually when i tried it, i couldn't halt or reboot my machine without an
> hard reset. I suppose that is a real example of bloated?
>
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=661239

That would be unreliable, not bloated.

Bloated could mean a number of things, roughly: Does too many things,
links to too many libraries, takes up too much memory, disk space or
possibly CPU time for what it does. Pretty much always a value
judgement at some level. Clearly all interrelated, but nothing
(directly) to do with unreliable. Doing a lot without good enough QC
could lead to unreliability, but it doesn't have to. And code can be
unreliable while being very tight and compact - though it should
decrease the likelihood.

Cheers,
Kelly Clowers

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/CAFoWM=92e1wXvvwgv4LjnSrg6OdRy5=hSzfp_3p2...@mail.gmail.com

[Salvo Tomaselli]

> > http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=661239
>
> Hmm, that doesn't look like a valid bug report to me. Especially I
> don't see why dhclient would be able to disrupt systemd in such a way
> that you'd need to do a hard reboot.

I don't see why either (that makes it a bug and not a super awesome feature i
suppose), but i couldn't reboot, maybe there were other causes for that but
the bugreport has just been ignored, and i am no expert on systemd and have no
idea where to begin to investigate.

Maybe systemd is faster, but i think being unable to do a normal reboot is an
important drawback.

--
Salvo Tomaselli

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/201211170022....@tiscali.it

[Paul Wise]

On Sat, Nov 17, 2012 at 7:22 AM, Salvo Tomaselli wrote:
>
>> > http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=661239
>>
>> Hmm, that doesn't look like a valid bug report to me. Especially I
>> don't see why dhclient would be able to disrupt systemd in such a way
>> that you'd need to do a hard reboot.
>
> I don't see why either (that makes it a bug and not a super awesome feature i
> suppose), but i couldn't reboot, maybe there were other causes for that but
> the bugreport has just been ignored, and i am no expert on systemd and have no
> idea where to begin to investigate.
>
> Maybe systemd is faster, but i think being unable to do a normal reboot is an
> important drawback.

systemd reboots just fine. It also kills all processes just like sysvinit does.

--
bye,
pabs

http://wiki.debian.org/PaulWise

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/CAKTje6E6b8Jj=JHKB+BBXZRKzLHNiri...@mail.gmail.com

[Salvo Tomaselli]

> systemd reboots just fine. It also kills all processes just like sysvinit
> does.

Some help in "how to find the cause" would be much more appreciated and useful
than: "it's not happening, you were just dreaming that".

--
Salvo Tomaselli

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/201211170221....@tiscali.it

[David Baird]

On Wed, Nov 14, 2012 at 3:37 AM, martin f krafft <mad...@debian.org> wrote:

also sprach Thomas Goirand <zi...@debian.org> [2012.11.14.0412 +0100]:

> As Gentoo guys and some major kernel people are protesting about the
> insanity Kay and Lennart have done to udev,

I cannot help but notice that Kay and Lennart were both
Gentoo-freaks when they took on udev and at least I always
attributed much of what was wrong with udev from the start (e.g. the
configuration file format) to being born in an environment where
people still compile from source. ;)

hahaha, funny. Not. These stupid jokes about gentoo are getting old now.

[Russell Coker]

On Sat, 17 Nov 2012, Paul Wise <pa...@debian.org> wrote:
> > Maybe systemd is faster, but i think being unable to do a normal reboot
> > is an important drawback.
>
> systemd reboots just fine. It also kills all processes just like sysvinit
> does.

I have also had problems with systemd not rebooting as fast as sysvinit and
sometimes hanging on a reboot or shutdown request. I haven't filed bug reports
because I haven't had enough time to investigate it properly (sorry, fixing
bugs in my own packages is my priority).

We should keep in mind that systemd is still relatively new and should be
expected to have some bugs.

Some of my virtual servers use sysvinit in Dom0 and systemd in the DomUs.
That gives the fast reboot for virtual servers that is convenience and the
reliability for the Dom0 which I haven't yet got from systemd.

--
My Main Blog http://etbe.coker.com.au/
My Documents Blog http://doc.coker.com.au/

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/201211171343....@coker.com.au

[Michael Biebl]

On 17.11.2012 03:43, Russell Coker wrote:
> On Sat, 17 Nov 2012, Paul Wise <pa...@debian.org> wrote:
>>> Maybe systemd is faster, but i think being unable to do a normal reboot
>>> is an important drawback.
>>
>> systemd reboots just fine. It also kills all processes just like sysvinit
>> does.
>
> I have also had problems with systemd not rebooting as fast as sysvinit and
> sometimes hanging on a reboot or shutdown request. I haven't filed bug reports

We have been seeing issues with dhcp/ifupdown hooks which were making
restart/reload requests for services during shutdown. This could lead to
90sec timeouts, before systemd killed those services.
This has been fixed in 44-5, so it would be great if you could test
again and see if it works now.

We are already way off-topic now, so please, if you still encounter a
problem, do file a bug.

Thanks,
Michael

--
Why is it that all of the instruments seeking intelligent life in the
universe are pointed away from Earth?

[Luke Leighton]

Roger Leigh <rleigh <at> codelibre.net> writes:

> If you want a reliable system, you need a reliable PID 1.

yes. this was i believe why richard lightman implemented depinit
in i think it was under 1,000 lines of code. he was delighted
when i came up with a simple modification which would allow
him to remove a further block of that, out-sourcing the job
to a shell script and increasing the flexibility at the same
time.

unfortunately i haven't heard from richard since.... 2008. he's
gone even more reclusive than he was before i met him. which
is a pity. we need more people in the world with his level
of quiet genius.

l.

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/loom.2012111...@post.gmane.org

[Marc Haber]

On Wed, 14 Nov 2012 15:04:35 +0100, John Paul Adrian Glaubitz
<glau...@physik.fu-berlin.de> wrote:
>One of the Arch developers actually made a couple of good points why
>they switched to systemd as default [1].

How many non-Linux platforms does Arch Linux (sic!) support?

I know of two non-Linux platforms that Debian prides itself in
supporting, which is the major stopper of systemd in Debian at the
moment.

Greetings
Marc
--
-------------------------------------- !! No courtesy copies, please !! -----
Marc Haber | " Questions are the | Mailadresse im Header
Mannheim, Germany | Beginning of Wisdom " | http://www.zugschlus.de/
Nordisch by Nature | Lt. Worf, TNG "Rightful Heir" | Fon: *49 621 72739834

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/E1TaXXY-...@swivel.zugschlus.de

[Paul Wise]

On Tue, Nov 20, 2012 at 4:01 AM, Marc Haber wrote:

> How many non-Linux platforms does Arch Linux (sic!) support?

Looks like just Hurd: http://www.archhurd.org/

It seems that they are talking about Arch BSD too:

http://bbs.archlinux.org/viewtopic.php?id=142679

> I know of two non-Linux platforms that Debian prides itself in
> supporting, which is the major stopper of systemd in Debian at the
> moment.

There is no reason for kFreeBSD and Hurd to stop Debian's Linux ports
from using systemd or upstart by default once wheezy is released. We
can keep sysvinit/openrc/busybox init/etc for kFreeBSD, Hurd and users
who need or prefer them.

--
bye,
pabs

http://wiki.debian.org/PaulWise

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/CAKTje6Hd410uikb2wEHz2+4L...@mail.gmail.com

[Marc Haber]

On Tue, 20 Nov 2012 10:13:28 +0800, Paul Wise <pa...@debian.org> wrote:
>There is no reason for kFreeBSD and Hurd to stop Debian's Linux ports
>from using systemd or upstart by default once wheezy is released. We
>can keep sysvinit/openrc/busybox init/etc for kFreeBSD, Hurd and users
>who need or prefer them.

Aren't the systemd makers trying hard to move existing functionality
from udev, consolekit, policykit and syslogd into systemd, effectively
making those unavailable for non-Linux?

Greetings
Marc
--
-------------------------------------- !! No courtesy copies, please !! -----
Marc Haber | " Questions are the | Mailadresse im Header
Mannheim, Germany | Beginning of Wisdom " | http://www.zugschlus.de/
Nordisch by Nature | Lt. Worf, TNG "Rightful Heir" | Fon: *49 621 72739834

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/E1TasUU-...@swivel.zugschlus.de

[Matthias Klumpp]

Hi!

2012/11/20 Marc Haber <mh+debi...@zugschlus.de>:

> On Tue, 20 Nov 2012 10:13:28 +0800, Paul Wise <pa...@debian.org> wrote:

>>[...]

> Aren't the systemd makers trying hard to move existing functionality
> from udev, consolekit, policykit and syslogd into systemd, effectively
> making those unavailable for non-Linux?

No, not really... ConsoleKit is replaced by a systemd-associated
daemon (logind) because of several design flaws of CK. This change
brought us e.g. full multiseat support.
Udev is already Linux-only, and it work fine without systemd.
PolicyKit has to stay separate, there are no plans in that direction -
it also would not make sense from an architectual point of view. But
systemd is using PolKit, of course.
For syslogd, systemd provides journald for those who want to use it,
but the Journal is no dependency of systemd. If applications use
existing syslog calls, nothing will change. For journald-specific
extra functionality (adding metadata), software can use journald-APIs,
which would make this software Linux-specific. But Lennart is thinking
about making this library work on non-Linux OSes too, which would
solve this "problem" too.
(Since most stuff uses syslog calls, this is not an issue. Also,
systemd does not require journald, even for Fedora it is not safe yet
if they will use the journal or not)
Cheers,
Matthias

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/CAKNHny_vVksTEwyPJ9sAUeZW...@mail.gmail.com

[Gergely Nagy]

Matthias Klumpp <matt...@tenstral.net> writes:

> For syslogd, systemd provides journald for those who want to use it,
> but the Journal is no dependency of systemd.

Wrong. You can't have any recent systemd without the Journal, and
'legacy' syslogds are piggybacking on /run/systemd/journal/syslog, where
the journal forwards syslog messages to. There's no way around that, you
can't turn the journal off.

> (Since most stuff uses syslog calls, this is not an issue. Also,
> systemd does not require journald, even for Fedora it is not safe yet
> if they will use the journal or not)

What Fedora & co are considering, is not requiring the journal (it comes
enabled since F17 already), but having no 'legacy' syslogd installed by
default along with it. [1]

[1]: http://fedoraproject.org/wiki/Features/systemd-journal

--
|8]

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/87r4nn2...@algernon.balabit

[Andrej N. Gritsenko]

Hello!

Gergely Nagy has written on Wednesday, 21 November, at 10:29:

>Matthias Klumpp <matt...@tenstral.net> writes:

>> For syslogd, systemd provides journald for those who want to use it,
>> but the Journal is no dependency of systemd.

>Wrong. You can't have any recent systemd without the Journal, and
>'legacy' syslogds are piggybacking on /run/systemd/journal/syslog, where
>the journal forwards syslog messages to. There's no way around that, you
>can't turn the journal off.

Does all that mean I cannot have /var/log/* logs with systemd? If
it's so then systemd should never come into any server installation at
the very least.

With best regards.
Andriy.

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121121144...@rep.kiev.ua

[Matthias Klumpp]

2012/11/21 Andrej N. Gritsenko <and...@rep.kiev.ua>:

> Hello!
>
> Gergely Nagy has written on Wednesday, 21 November, at 10:29:
>>Matthias Klumpp <matt...@tenstral.net> writes:
>
>>> For syslogd, systemd provides journald for those who want to use it,
>>> but the Journal is no dependency of systemd.
>>Wrong. You can't have any recent systemd without the Journal, and
>>'legacy' syslogds are piggybacking on /run/systemd/journal/syslog, where
>>the journal forwards syslog messages to. There's no way around that, you
>>can't turn the journal off.

Really? I'll take a look at this again, last time I checked journald
was not mandatory. Sorry for the misinformation then!

> Does all that mean I cannot have /var/log/* logs with systemd? If
> it's so then systemd should never come into any server installation at
> the very least.

No, it just means that you have two log systems running.

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/CAKNHny_F9XSB9USgXec9QSiX...@mail.gmail.com

[Gergely Nagy]

"Andrej N. Gritsenko" <and...@rep.kiev.ua> writes:

> Hello!
>
> Gergely Nagy has written on Wednesday, 21 November, at 10:29:
>>Matthias Klumpp <matt...@tenstral.net> writes:
>
>>> For syslogd, systemd provides journald for those who want to use it,
>>> but the Journal is no dependency of systemd.
>
>>Wrong. You can't have any recent systemd without the Journal, and
>>'legacy' syslogds are piggybacking on /run/systemd/journal/syslog, where
>>the journal forwards syslog messages to. There's no way around that, you
>>can't turn the journal off.
>
> Does all that mean I cannot have /var/log/* logs with systemd? If
> it's so then systemd should never come into any server installation at
> the very least.

If you read my mail further, your question is answered there. But I'll
repeat it, for good measures: no, it does not mean that. I only means
that your syslogd of choice is getting its input from
/run/systemd/journal/syslog instead of /dev/log.

--
|8]

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/87obiqa...@luthien.mhp

[Andrej N. Gritsenko]

Hello!

Gergely Nagy has written on Thursday, 22 November, at 0:05:

>If you read my mail further, your question is answered there. But I'll
>repeat it, for good measures: no, it does not mean that. I only means
>that your syslogd of choice is getting its input from
>/run/systemd/journal/syslog instead of /dev/log.

Thank you very much!

Cheers!
Andriy.

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121122123...@rep.kiev.ua

[Toni Mueller]

Hi,

On Fri, Nov 16, 2012 at 02:09:51AM +0800, Thomas Goirand wrote:
> On 11/14/2012 11:12 AM, Thomas Goirand wrote:
> > The full thread is here:
> > http://thread.gmane.org/gmane.linux.gentoo.project/2262

> This thread was originally about udev, yet everyone is starting again the
> systemd / upstart / sysv-rc war. I think we can agree that we don't about

I, for one, wholeheartedly welcome the fork, as I hope that this will
help getting back some of the modularity in Linux that was there, once
upon a time, and which contributed to making Linux a robust platform -
sometime in the past.


Kind regards,
--Toni++

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121124155...@spruce.wiehl.oeko.net

[John Paul Adrian Glaubitz]

On Sat, Nov 24, 2012 at 04:58:04PM +0100, Toni Mueller wrote:
> > This thread was originally about udev, yet everyone is starting again the
> > systemd / upstart / sysv-rc war. I think we can agree that we don't about
>
> I, for one, wholeheartedly welcome the fork, as I hope that this will
> help getting back some of the modularity in Linux that was there, once
> upon a time, and which contributed to making Linux a robust platform -
> sometime in the past.

Well, while I'm usually not against forks - I like MATE for example
very much - I don't think this fork is going anywhere soon. I have
followed their development a bit and read the discussion involving
Greg Kroah-Hartman and others and their comments on the fork and I
don't think the eudev people know what they're doing.

They're constantly claiming, for example, that udev and systemd break
a separate /usr partition which is simply not true. systemd just warns
you about it. Furthermore, they're randomly removing code from udev
which they don't seem to understand which means they will probably
break something some time.

I don't think these guys have the expertise to work on a udev fork (I
wouldn't claim that for myself either). Just look at their discussion
about creating a free BIOS replacement [1], it's ridiculous.

Coming back to your original complaint. The discussion about init
systems naturally came about because this is actually the reason udev
was forked by Gentoo in the first place. They want to stick to their
init solution OpenRC - no matter what - like Ubuntu wants to stick to
upstart.

If both Ubuntu and Gentoo would just go with the rest of the community
and accept systemd, we wouldn't have to bother whether udev runs
without systemd or not. I don't see anyway why something as low-level
as udev should be highly portable in the first place.

Cheers,

Adrian

> [1] https://groups.google.com/forum/?hl=en&fromgroups=#!topic/linux.gentoo.dev/yWQGqxsfjI0

--
.''`. John Paul Adrian Glaubitz
: :' : Debian Developer - glau...@debian.org
`. `' Freie Universitaet Berlin - glau...@physik.fu-berlin.de
`- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/2012112416...@physik.fu-berlin.de

[Toni Mueller]

On Sat, Nov 24, 2012 at 05:15:25PM +0100, John Paul Adrian Glaubitz wrote:
> If both Ubuntu and Gentoo would just go with the rest of the community
> and accept systemd, we wouldn't have to bother whether udev runs
> without systemd or not.

I would highly prefer a system where I can take small bites if I want
to, and where components are as portable as possible, and as it stands,
I am very uncomfortable with systemd, too. It's certainly
interesting, but having a hard dependency on it is imho a no-no.

Please take a step back, look at the bigger picture, and tell me
whether you think that we, as Debianites, or as Free Software
Advocates, are still heading in the right direction:

1. What does it mean if more and more software is required to run a
Linux system, in the various scenarios, and losing the ability to
swap components w/o major kernel hackery? So far, being highly
modular was a way to contain the complexity, and finish off bugs.
It's not only "more eyeballs" to catch them, it's also the complexity
that makes it increasingly hard for people to understand what's going
on in the first place, so the problem of having the required
knowledge, that you highlighted in your message, will only get worse
and worse with tighter integration.

2. What does it mean if more and more software only runs on Linux?

3. What does it mean that - my claim/experience - more and more Linux
software is simply broken (see Gnome for a popular example, but I
have more)?

> I don't see anyway why something as low-level
> as udev should be highly portable in the first place.

Maybe, but I do see why systemd must not be a hard dependency of the
Linux kernel, and if systemd is basically the user-space part of udev,
and both can't live without each other, then something is fundamentally
wrong in the design. IMHO.



Kind regards,
--Toni++

PS: Please don't Cc' me, I'm on the list.

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121124170...@spruce.wiehl.oeko.net

[John Paul Adrian Glaubitz]

On Sat, Nov 24, 2012 at 06:03:02PM +0100, Toni Mueller wrote:
> On Sat, Nov 24, 2012 at 05:15:25PM +0100, John Paul Adrian Glaubitz wrote:
> > If both Ubuntu and Gentoo would just go with the rest of the community
> > and accept systemd, we wouldn't have to bother whether udev runs
> > without systemd or not.
>
> I would highly prefer a system where I can take small bites if I want
> to, and where components are as portable as possible

Why? Why would you want to rip such low-level stuff apart? It
seriously doesn't make any sense unless you need a highly-customized
setup, for embedded applications, for example.

> and as it stands,
> I am very uncomfortable with systemd, too. It's certainly
> interesting, but having a hard dependency on it is imho a no-no.

Again. What's so bad with systemd? I don't get it. I really don't get
it. You're not seeing your init daemon 99% of the time you're using
your computer, so why even bother with it? When you have something
such low-level, you're best off with taking the best solution which
is clearly systemd and which is why most distributions are adopting
it.

> Please take a step back, look at the bigger picture, and tell me
> whether you think that we, as Debianites, or as Free Software
> Advocates, are still heading in the right direction:
>
> 1. What does it mean if more and more software is required to run a
> Linux system, in the various scenarios, and losing the ability to
> swap components w/o major kernel hackery?

Because it's a LOW-LEVEL component and you don't profit at all by
choosing any different solution. It's not like your browser or desktop
which you can choose on your personal preferences. Argueing which init
daemon is the best is like argueing over which oil pump in your car to
choose based on preferences. You choose the one that does the job
best, it doesn't have to be customizable, that's just non-sense.

> So far, being highly
> modular was a way to contain the complexity, and finish off bugs.

So, you think code is more maintainable when it's spread across dozens
of projects instead of doing the work in one repository? Why should we
have separate repositories for atd, crond, anacron, xinetd, init, rc,
watchdogd, autofs and so on when these projects are more or less
something that should be performed by ONE daemon, because what these
daemons do overlaps quite a lot?

> It's not only "more eyeballs" to catch them

systemd has over 118 contributors, that makes 236 eyeballs. I guess
that's much more than any eyeballs in any of the traditional daemons
combined.

> it's also the complexity
> that makes it increasingly hard for people to understand what's going
> on in the first place, so the problem of having the required
> knowledge, that you highlighted in your message, will only get worse
> and worse with tighter integration.

systemd is not complex, sysvinit is. Just compare the average init
script with a systemd unit file and you will understand.

> 2. What does it mean if more and more software only runs on Linux?

It's a very good thing. It makes Linux stronger as a free platform
which is what we need in order to be able to compete against the
proprietary platforms. We need to combine power.

You probably want to hear now that's unfair to the other free
operating systems, but seriously, I don't care. You want to dictate
developers of free software not to take advantage of Linux-specific
features. You want to force them to consider stuff like FreeBSD as
well.

> 3. What does it mean that - my claim/experience - more and more Linux
> software is simply broken (see Gnome for a popular example, but I
> have more)?

It's broken because it doesn't run on your favourite non-Linux
operating system? I don't think so.

> > I don't see anyway why something as low-level
> > as udev should be highly portable in the first place.
>
> Maybe, but I do see why systemd must not be a hard dependency of the
> Linux kernel, and if systemd is basically the user-space part of udev,
> and both can't live without each other, then something is fundamentally
> wrong in the design. IMHO.

systemd has a hard dependency on the Linux kernel because it takes
advantage of many Linux-specific features and it would be stupid not
to do that. Linux has these features, so we should be free to use
them when we can.

Do the FreeBSD people develop all their stuff in a fashion that it
would run on Linux as well?

I'm sorry for the harsh tone, but it's really something that annoys
me, people constantly complaining about systemd but never really
coming up with good arguments why something as low-level as
systemd/udev should be replacable in the first place. 95% of the users
don't care and just want something that's reliable.

Adrian

--
.''`. John Paul Adrian Glaubitz
: :' : Debian Developer - glau...@debian.org
`. `' Freie Universitaet Berlin - glau...@physik.fu-berlin.de
`- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/2012112417...@physik.fu-berlin.de

[Russ Allbery]

John Paul Adrian Glaubitz <glau...@physik.fu-berlin.de> writes:
> On Sat, Nov 24, 2012 at 04:58:04PM +0100, Toni Mueller wrote:

>> I, for one, wholeheartedly welcome the fork, as I hope that this will
>> help getting back some of the modularity in Linux that was there, once
>> upon a time, and which contributed to making Linux a robust platform -
>> sometime in the past.

> Well, while I'm usually not against forks - I like MATE for example
> very much - I don't think this fork is going anywhere soon. I have
> followed their development a bit and read the discussion involving
> Greg Kroah-Hartman and others and their comments on the fork and I
> don't think the eudev people know what they're doing.

[...]

I really wish people would stop having this debate.

It is completely pointless for us to argue here over whether or not the
fork will be successful. The outcome of that argument is completely
irrelevant to the world: even if we all decide that the fork will be
successful or all decide that the fork will be unsuccessful, it will not
have the slightest effect on reality. Meanwhile, it will become obvious
(or at least much more obvious) whether the fork is successful if we just
wait and see what happens.

All that debating its possible success is doing is hardening everyone's
positions (about something for which there's no point in having a
position!) and creating hard feelings.

For those of us who are not directly involved in upstream kernel
development or other affected upstream projects, the *only* thing that
anyone has to worry about at this very, very early stage is whether you,
personally, want to go help with the fork. If the answer is no, then all
you have to do right now is watch and see.

Let's please not turn this into a sporting event where we stand on the
sidelines and root for our "side" instead of wait to evaluate things on
their merits. There's too much of that in free software already.

--
Russ Allbery (r...@debian.org) <http://www.eyrie.org/~eagle/>

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/87txsex...@windlord.stanford.edu

[Carlos Alberto Lopez Perez]

FYI, Yet another episode of the Linux init drama:


https://wiki.ubuntu.com/FoundationsTeam/Specs/RaringUpstartUserSessions

https://plus.google.com/115547683951727699051/posts/ZZWLtq6tYdn

[Steve McIntyre]

Adrian wrote:
>
>If both Ubuntu and Gentoo would just go with the rest of the community
>and accept systemd, we wouldn't have to bother whether udev runs
>without systemd or not.

Please drop the systemd propaganda crap. We get enough of that from
Lennart already.

--
Steve McIntyre, Cambridge, UK. st...@einval.com
"We're the technical experts. We were hired so that management could
ignore our recommendations and tell us how to do our jobs." -- Mike Andrews

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/E1TcKPz-...@mail.einval.com

[Andrej N. Gritsenko]

Hello!

Carlos Alberto Lopez Perez has written on Saturday, 24 November, at 19:20:

>FYI, Yet another episode of the Linux init drama:


>https://wiki.ubuntu.com/FoundationsTeam/Specs/RaringUpstartUserSessions

>https://plus.google.com/115547683951727699051/posts/ZZWLtq6tYdn

It is only me who thinks that it is going by Windows steps? With only
difference Windows has more manpower to manage that bloated monster bugs.

WBR, Andriy.

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121124184...@rep.kiev.ua

[John Paul Adrian Glaubitz]

On Sat, Nov 24, 2012 at 07:20:02PM +0100, Carlos Alberto Lopez Perez wrote:
> https://wiki.ubuntu.com/FoundationsTeam/Specs/RaringUpstartUserSessions

This is actually going to be very interesting to see if they are able
to extend upstart in such a way that they can use it for session
management similar to systemd-loginctl.

Adrian

--
.''`. John Paul Adrian Glaubitz
: :' : Debian Developer - glau...@debian.org
`. `' Freie Universitaet Berlin - glau...@physik.fu-berlin.de
`- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/2012112419...@physik.fu-berlin.de

[John Paul Adrian Glaubitz]

On Sat, Nov 24, 2012 at 08:17:35PM +0100, John Paul Adrian Glaubitz wrote:
> On Sat, Nov 24, 2012 at 07:20:02PM +0100, Carlos Alberto Lopez Perez wrote:
> > https://wiki.ubuntu.com/FoundationsTeam/Specs/RaringUpstartUserSessions
>
> This is actually going to be very interesting to see if they are able
> to extend upstart in such a way that they can use it for session
> management similar to systemd-loginctl.

While reading a bit on it, I found this passage:

> By making use of a Linux-specific prctl(2) call, we effectively tie
> Upstart to systems running with a Linux kernel. This is a major
> restriction, but porting to other systems is already complicated by
> the fact that even the BSDs do not provide a full POSIX environment
> (missing "waitid(2)" for example).

So it's not just systemd which runs into the situation that at some
point they have to drop support for non-Linux kernels because they
need a Linux-specific feature, in this case prctl.

[Russ Allbery]

John Paul Adrian Glaubitz <glau...@physik.fu-berlin.de> writes:

> On Sat, Nov 24, 2012 at 08:17:35PM +0100, John Paul Adrian Glaubitz wrote:

>> By making use of a Linux-specific prctl(2) call, we effectively tie
>> Upstart to systems running with a Linux kernel. This is a major
>> restriction, but porting to other systems is already complicated by the
>> fact that even the BSDs do not provide a full POSIX environment
>> (missing "waitid(2)" for example).

> So it's not just systemd which runs into the situation that at some
> point they have to drop support for non-Linux kernels because they need
> a Linux-specific feature, in this case prctl.

This is not a revelation. We've known this for, literally, years. It's
been part of every previous init script discussion we've had on
debian-devel. However, many of those features can be made optional or be
implemented another way, not to mention that the FreeBSD kernel is not
static either and can adapt and adopt new features itself.

Porting any more modern, event-driven, session-aware init system to
FreeBSD is going to require actual work. I don't think anyone is under
any illusions to the contrary.

--
Russ Allbery (r...@debian.org) <http://www.eyrie.org/~eagle/>

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/87624ux...@windlord.stanford.edu

[Guillem Jover]

On Sat, 2012-11-24 at 20:29:51 +0100, John Paul Adrian Glaubitz wrote:
> While reading a bit on it, I found this passage:
>

> On Sat, Nov 24, 2012 at 08:17:35PM +0100, John Paul Adrian Glaubitz wrote:
> > By making use of a Linux-specific prctl(2) call, we effectively tie
> > Upstart to systems running with a Linux kernel. This is a major
> > restriction, but porting to other systems is already complicated by
> > the fact that even the BSDs do not provide a full POSIX environment
> > (missing "waitid(2)" for example).
>
> So it's not just systemd which runs into the situation that at some
> point they have to drop support for non-Linux kernels because they
> need a Linux-specific feature, in this case prctl.

waitid is supported on FreeBSD. Regarding prctl(2), it seems upstart
is supposed to handle Linux kernels w/o prctl(PR_SET_CHILD_SUBREAPER)
support gracefully by printing a warning and just ignoring respawn
statements, so the same could be applied to non-Linux kernels. Of
course there's other things that might need porting currently.

Regards,
Guillem

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121124212...@gaara.hadrons.org

[John Paul Adrian Glaubitz]

On Sat, Nov 24, 2012 at 10:28:46PM +0100, Guillem Jover wrote:
>
> waitid is supported on FreeBSD.

Are you sure? According to their status page [1] it's not yet fully
implemented. The page is dated to last October.

Adrian

> [1] http://www.freebsd.org/projects/c99/index.html

--
.''`. John Paul Adrian Glaubitz
: :' : Debian Developer - glau...@debian.org
`. `' Freie Universitaet Berlin - glau...@physik.fu-berlin.de
`- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/2012112421...@physik.fu-berlin.de

[Josselin Mouette]

Le samedi 24 novembre 2012 à 18:25 +0000, Steve McIntyre a écrit :
> Please drop the systemd propaganda crap. We get enough of that from
> Lennart already.

OTOH we also get quite enough of FUD from people who don’t know what
systemd is but don’t want us to use it.

--
.''`. Josselin Mouette
: :' :
`. `'
`-

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/1353798957.4220.23.camel@pi0307572

[Guillem Jover]

On Sat, 2012-11-24 at 22:46:29 +0100, John Paul Adrian Glaubitz wrote:
> On Sat, Nov 24, 2012 at 10:28:46PM +0100, Guillem Jover wrote:
> > waitid is supported on FreeBSD.
>
> Are you sure? According to their status page [1] it's not yet fully
> implemented. The page is dated to last October.

<https://github.com/freebsd/freebsd/commit/a31680b74e1dc8fb7acf515826812d0237aba661>
<https://github.com/freebsd/freebsd/commit/6da05143f33f9398d99a3423476dbe49b43a98a6>
<https://github.com/freebsd/freebsd/commit/31050a67fbf31c3de5e21c98ea1520c6c204496e>

Regards,
Guillem

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121124232...@gaara.hadrons.org

[Steve Langasek]

On Sat, Nov 24, 2012 at 08:29:51PM +0100, John Paul Adrian Glaubitz wrote:
> On Sat, Nov 24, 2012 at 08:17:35PM +0100, John Paul Adrian Glaubitz wrote:
> > On Sat, Nov 24, 2012 at 07:20:02PM +0100, Carlos Alberto Lopez Perez wrote:
> > > https://wiki.ubuntu.com/FoundationsTeam/Specs/RaringUpstartUserSessions

> > This is actually going to be very interesting to see if they are able
> > to extend upstart in such a way that they can use it for session
> > management similar to systemd-loginctl.

> While reading a bit on it, I found this passage:

> > By making use of a Linux-specific prctl(2) call, we effectively tie
> > Upstart to systems running with a Linux kernel. This is a major
> > restriction, but porting to other systems is already complicated by
> > the fact that even the BSDs do not provide a full POSIX environment
> > (missing "waitid(2)" for example).

> So it's not just systemd which runs into the situation that at some
> point they have to drop support for non-Linux kernels because they
> need a Linux-specific feature, in this case prctl.

Upstart is already not portable to non-Linux kernels. This is a known
issue; had it not been for sensitivity to not breaking Debian's non-Linux
ports, it's likely that upstart would have been the default in Debian before
systemd was a glimmer in Lennart's eye.

Unlike systemd, upstart is open to being ported to non-Linux kernels.
Sadly, so far no one has stepped up to do this work.

--
Steve Langasek Give me a lever long enough and a Free OS
Debian Developer to set it on, and I can move the world.
Ubuntu Developer http://www.debian.org/
slan...@ubuntu.com vor...@debian.org

[Thomas Goirand]

On 11/25/2012 12:15 AM, John Paul Adrian Glaubitz wrote:
> They're constantly claiming, for example, that udev and systemd break
> a separate /usr partition which is simply not true.

I believe you've been reading too much L. Poettring. Yes, lots of
udev stuff are moving to /usr, and this is a fact. Yes, lots of
things are annoying in the merge for someone who wishes to use
udev alone, and not systemd. That is a fact as well.

It would help the discussion if there was nobody always claiming
the opposite.

On 11/25/2012 12:15 AM, John Paul Adrian Glaubitz wrote:
> If both Ubuntu and Gentoo would just go with the rest of the community
> and accept systemd

Yeah, right! Along the line with "why are these idiots so vocal". That is
in fact one of the main concern about udev/systemd people: they don't
care about others, refuse patches, and always claim others are stupid.
You shouldn't go through this dangerous path as well, IMO.

Thomas

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/50B1A41...@debian.org

[John Paul Adrian Glaubitz]

On Sun, Nov 25, 2012 at 12:52:47PM +0800, Thomas Goirand wrote:
> On 11/25/2012 12:15 AM, John Paul Adrian Glaubitz wrote:
> > They're constantly claiming, for example, that udev and systemd break
> > a separate /usr partition which is simply not true.
> I believe you've been reading too much L. Poettring.

And again the same idiotic and childish style of argumentation. It has
nothing to do with Poettering. Please come up with real arguments
instead of just saying systemd is broken and the author is an idiot.

> Yes, lots of
> udev stuff are moving to /usr, and this is a fact. Yes, lots of
> things are annoying in the merge for someone who wishes to use
> udev alone, and not systemd. That is a fact as well.

There is tons of stuff that you would need to fix, not just udev
stuff. See [1].

> It would help the discussion if there was nobody always claiming
> the opposite.

What? I'm not claiming the opposite, I am claiming facts. Is it false
that all these binaries mentioned in [1] are required for early boot
up and are located in /usr? Just check it yourself.

> On 11/25/2012 12:15 AM, John Paul Adrian Glaubitz wrote:
> > If both Ubuntu and Gentoo would just go with the rest of the community
> > and accept systemd
> Yeah, right! Along the line with "why are these idiots so vocal". That is
> in fact one of the main concern about udev/systemd people: they don't
> care about others, refuse patches, and always claim others are stupid.
> You shouldn't go through this dangerous path as well, IMO.

Sorry, but I wouldn't touch code with a ten-feet pole who from someone
is so naive claiming that he knows more about writing an open source
BIOS than the people at Coreboot who have been doing that since
1999. I started right out laughing when he claiming Coreboot is a
Linux distribution and starting a x86 computer just involves setting
up a few interrupt vectors.

I know several people in the Coreboot project personal and the
claims that Richard makes in [2] are just ridiculous and show the same
attitude he has towards working on udev.

Even Greg is ridiculing their work [3]. And he should know, he came up
with udev in the first place and he is even a Gentoo developer. Don't
you think he understands you a little more than anyone else, so he can
assess when people do the right things with the code or not?

Adrian

> [1] http://freedesktop.org/wiki/Software/systemd/separate-usr-is-broken
> [2] https://groups.google.com/forum/?hl=en&fromgroups=#!topic/linux.gentoo.dev/yWQGqxsfjI0
> [3] http://thread.gmane.org/gmane.linux.gentoo.project/2226/focus=81216

--
.''`. John Paul Adrian Glaubitz
: :' : Debian Developer - glau...@debian.org
`. `' Freie Universitaet Berlin - glau...@physik.fu-berlin.de
`- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121125073...@physik.fu-berlin.de

[John Paul Adrian Glaubitz]

On Sun, Nov 25, 2012 at 08:35:22AM +0100, John Paul Adrian Glaubitz wrote:
> Sorry, but I wouldn't touch code with a ten-feet pole who from someone
> is so naive claiming that he knows more about writing an open source
> BIOS than the people at Coreboot who have been doing that since
> 1999. I started right out laughing when he claiming Coreboot is a
> Linux distribution and starting a x86 computer just involves setting
> up a few interrupt vectors.
>
> I know several people in the Coreboot project personal and the
> claims that Richard makes in [2] are just ridiculous and show the same
> attitude he has towards working on udev.
>
> Even Greg is ridiculing their work [3]. And he should know, he came up
> with udev in the first place and he is even a Gentoo developer. Don't
> you think he understands you a little more than anyone else, so he can
> assess when people do the right things with the code or not?

And if that's not already enough, they even started removing copyright
information from the udev files headers [1] which I find rude and
disrespectful. They claim that they can do that because it's free
software, after all. Again, it just shows they have no idea what
they're talking about.

Going with this philosophy, their code wouldn't probably even meet the
Debian Free Software Guidelines (DFSG). Removing copyright information
of other people's work is a no no.

Adrian

> [1] http://thread.gmane.org/gmane.linux.gentoo.project/2226/focus=81216

--
.''`. John Paul Adrian Glaubitz
: :' : Debian Developer - glau...@debian.org
`. `' Freie Universitaet Berlin - glau...@physik.fu-berlin.de
`- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913


--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121125075...@physik.fu-berlin.de

[John Paul Adrian Glaubitz]

On Sun, Nov 25, 2012 at 10:48:27PM +1300, Chris Bannister wrote:

> https://lists.linuxaudio.org/mailarchive/lau/2012/11/21/194431
>
> There is a rather bad smell regarding all this.

None of the systemd advocates ever mentioned for example the real
reason why it uses such an ugly configuration syntax and Windoze
format files. ^^^^^^^
I stopped reading
there

Adrian

--
.''`. John Paul Adrian Glaubitz
: :' : Debian Developer - glau...@debian.org
`. `' Freie Universitaet Berlin - glau...@physik.fu-berlin.de
`- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121125100...@physik.fu-berlin.de

[Chris Bannister]

On Sat, Nov 24, 2012 at 08:43:07PM +0200, Andrej N. Gritsenko wrote:
> Hello!
>
> Carlos Alberto Lopez Perez has written on Saturday, 24 November, at 19:20:
> >FYI, Yet another episode of the Linux init drama:
>
>
> >https://wiki.ubuntu.com/FoundationsTeam/Specs/RaringUpstartUserSessions
>
> >https://plus.google.com/115547683951727699051/posts/ZZWLtq6tYdn
>
>
> It is only me who thinks that it is going by Windows steps? With only
> difference Windows has more manpower to manage that bloated monster bugs.

https://lists.linuxaudio.org/mailarchive/lau/2012/11/21/194431

There is a rather bad smell regarding all this.

--
"If you're not careful, the newspapers will have you hating the people
who are being oppressed, and loving the people who are doing the
oppressing." --- Malcolm X

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121125094827.GI9596@tal

[Thomas Goirand]

Hi,

First, I'm registered to the list. So please *do not* Cc: me.

On 11/25/2012 03:35 PM, John Paul Adrian Glaubitz wrote:
> On Sun, Nov 25, 2012 at 12:52:47PM +0800, Thomas Goirand wrote:
>> On 11/25/2012 12:15 AM, John Paul Adrian Glaubitz wrote:
>>> They're constantly claiming, for example, that udev and systemd break
>>> a separate /usr partition which is simply not true.
>> I believe you've been reading too much L. Poettring.
>
> And again the same idiotic and childish style of argumentation. It has
> nothing to do with Poettering. Please come up with real arguments
> instead of just saying systemd is broken and the author is an idiot.

I do just right after, giving you the real argument. This was just an
introduction, and I didn't expect that you would take it that aggressively.

On your side, please refrain from:
1/ quoting only the introduction sentence just right before the real
argument
2/ and then complain I have none, with your quoting style fooling the
readers of the thread.

If you want real arguments, then fight them rather than the decorative
sentences around them.

Besides this, I have *never* wrote that the authors of systemd are
idiots. *NEVER* Please don't put words into my mouth.

Instead, I wrote that upstream authors think others are idiots (it shows
in various threads). That's very different.

>> Yes, lots of
>> udev stuff are moving to /usr, and this is a fact. Yes, lots of
>> things are annoying in the merge for someone who wishes to use
>> udev alone, and not systemd. That is a fact as well.
>
> There is tons of stuff that you would need to fix, not just udev
> stuff. See [1].

The point is, systemd and udev have recently been patched by upstream so
that things are going *even more* on the direction of having stuff
stored in /usr.

That is exactly this kind of things (and others, like merging systemd
and udev), going to the wrong direction, which pushed Gentoo people to fork.

I am *not* giving *my* opinion here, I'm not telling who's right or
wrong, just facts.

>> It would help the discussion if there was nobody always claiming
>> the opposite.
>
> What? I'm not claiming the opposite, I am claiming facts.

See above ...

>> On 11/25/2012 12:15 AM, John Paul Adrian Glaubitz wrote:
>>> If both Ubuntu and Gentoo would just go with the rest of the community
>>> and accept systemd
>> Yeah, right! Along the line with "why are these idiots so vocal". That is
>> in fact one of the main concern about udev/systemd people: they don't
>> care about others, refuse patches, and always claim others are stupid.
>> You shouldn't go through this dangerous path as well, IMO.
>
> Sorry, but I wouldn't touch code with a ten-feet pole who from someone
> is so naive claiming that he knows more about writing an open source
> BIOS than the people at Coreboot who have been doing that since
> 1999. I started right out laughing when he claiming Coreboot is a
> Linux distribution and starting a x86 computer just involves setting
> up a few interrupt vectors.
>
> I know several people in the Coreboot project personal and the
> claims that Richard makes in [2] are just ridiculous and show the same
> attitude he has towards working on udev.
>
> Even Greg is ridiculing their work [3]. And he should know, he came up
> with udev in the first place and he is even a Gentoo developer. Don't
> you think he understands you a little more than anyone else, so he can
> assess when people do the right things with the code or not?

This is the kind of aggressive attitude which happened with systemd
upstream, claiming that everyone are idiots, while in fact they've been
pointing rightly at problems. You just made a new occurrence of it. :(

Thomas

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/50B2283B...@debian.org

[John Paul Adrian Glaubitz]

On Sun, Nov 25, 2012 at 10:16:27PM +0800, Thomas Goirand wrote:
> >> Yes, lots of
> >> udev stuff are moving to /usr, and this is a fact. Yes, lots of
> >> things are annoying in the merge for someone who wishes to use
> >> udev alone, and not systemd. That is a fact as well.
> >
> > There is tons of stuff that you would need to fix, not just udev
> > stuff. See [1].
>
> The point is, systemd and udev have recently been patched by upstream so
> that things are going *even more* on the direction of having stuff
> stored in /usr.

Which is still not really a problem when tons of other daemons have
done the same already. We're not "fixing" this by patching systemd and
udev when many other daemons behave the same way. So, it's pointless
anyway.

Besides, can you elaborate what is so important in having /usr
separate? I see that it made sense back on the old Unix workstations
where you could split partitions across different disks, but I don't
see the point nowadays where a cheap harddisk has 1TB of space.

> That is exactly this kind of things (and others, like merging systemd
> and udev), going to the wrong direction, which pushed Gentoo people
> to fork.

Which I don't agree. Just look at FreeBSD, they have the whole
operating system core merged with their kernel source. I don't see
anyone complaining there either.

> >> On 11/25/2012 12:15 AM, John Paul Adrian Glaubitz wrote:
> >>> If both Ubuntu and Gentoo would just go with the rest of the community
> >>> and accept systemd
> >> Yeah, right! Along the line with "why are these idiots so vocal". That is
> >> in fact one of the main concern about udev/systemd people: they don't
> >> care about others, refuse patches, and always claim others are stupid.
> >> You shouldn't go through this dangerous path as well, IMO.
> >
> > Sorry, but I wouldn't touch code with a ten-feet pole who from someone
> > is so naive claiming that he knows more about writing an open source
> > BIOS than the people at Coreboot who have been doing that since
> > 1999. I started right out laughing when he claiming Coreboot is a
> > Linux distribution and starting a x86 computer just involves setting
> > up a few interrupt vectors.
> >
> > I know several people in the Coreboot project personal and the
> > claims that Richard makes in [2] are just ridiculous and show the same
> > attitude he has towards working on udev.
> >
> > Even Greg is ridiculing their work [3]. And he should know, he came up
> > with udev in the first place and he is even a Gentoo developer. Don't
> > you think he understands you a little more than anyone else, so he can
> > assess when people do the right things with the code or not?
>
> This is the kind of aggressive attitude which happened with systemd
> upstream, claiming that everyone are idiots, while in fact they've been
> pointing rightly at problems. You just made a new occurrence of it. :(

And you are dodging my questions. Don't you think that people like
Richard who make such bold claims that writing a free BIOS replacement
is a matter of a few hundred lines of assembly, ridiculing the almost
15 years of development of Coreboot, cannot be taken seriously?

Do you think that Greg - being a Gentoo developer himself - would make
such statements if he didn't knew what he was talking about?

Sorry, but no. I have seen such claims in the past very often where
people ridiculed existing software stacks and claiming they could do
something better in no time [1]. This guy would even remain on his
bold point of view after one of the Ardour and Jack developers
explained him why he was wrong [2].

I didn't mean to be aggressive in any way, I'm just annoyed. Please
don't get me wrong about forks, they are just fine. As I said before,
I like MATE very much, for example, because these people do great work
and they even collaborate with the original authors from GNOME.

But I don't support forks which are created based on false assumptions
of the original software. Another bad example where people forked and
obviously don't know what they're doing is Trinity [3].

Adrian

> [1] http://klang.eudyptula.org/
> [2] http://ardour.org/pd_on_klang
> [3] http://blog.martin-graesslin.com/blog/2012/10/maintaining-history-done-wrong/

--
.''`. John Paul Adrian Glaubitz
: :' : Debian Developer - glau...@debian.org
`. `' Freie Universitaet Berlin - glau...@physik.fu-berlin.de
`- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121125144...@physik.fu-berlin.de

[Thomas Goirand]

On 11/25/2012 01:30 AM, John Paul Adrian Glaubitz wrote:
> Why? Why would you want to rip such low-level stuff apart?

Well, isn't it the opposite thing that is happening? "Such low-level
stuff" are being merged (with systemd+udev merge), they were
separated projects before.

So, I'd rather ask you: why would you want "such low-level stuff"
to merge, since some others like it separated (like for example,
to be able to have the choice of replacing one or another)?

Thomas

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/50B230CA...@debian.org

[Thomas Goirand]

On 11/25/2012 02:19 AM, Russ Allbery wrote:
> I really wish people would stop having this debate.
>
> It is completely pointless for us to argue here over whether or not the
> fork will be successful. The outcome of that argument is completely
> irrelevant to the world: even if we all decide that the fork will be
> successful or all decide that the fork will be unsuccessful, it will not
> have the slightest effect on reality. Meanwhile, it will become obvious
> (or at least much more obvious) whether the fork is successful if we just
> wait and see what happens.
>
> All that debating its possible success is doing is hardening everyone's
> positions (about something for which there's no point in having a
> position!) and creating hard feelings.
>
> For those of us who are not directly involved in upstream kernel
> development or other affected upstream projects, the *only* thing that
> anyone has to worry about at this very, very early stage is whether you,
> personally, want to go help with the fork.

I probably would have like to at least contribute a bit what I can
(I don't think I could have helped a lot with major core stuff on a
udev fork, but I'm sure there are other areas where help would
be welcome). But if everyone in Debian already rejects it before
we have anything to show, then I wont waste my time.

In fact, seeing how this discussion turns out, I probably even
regret even the few hours I spent on a Sunday, few months ago,
trying to build OpenRC on Debian.

Thomas

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/50B23402...@debian.org

[John Paul Adrian Glaubitz]

On Sun, Nov 25, 2012 at 10:52:58PM +0800, Thomas Goirand wrote:
> On 11/25/2012 01:30 AM, John Paul Adrian Glaubitz wrote:
> > Why? Why would you want to rip such low-level stuff apart?
>
> Well, isn't it the opposite thing that is happening? "Such low-level
> stuff" are being merged (with systemd+udev merge), they were
> separated projects before.
>
> So, I'd rather ask you: why would you want "such low-level stuff"
> to merge, since some others like it separated (like for example,
> to be able to have the choice of replacing one or another)?

Well, systemd and udev are developed by the same developers. Both
daemons interact very closely and integration of the sources was the
natural consequence.

Yes, it makes it more difficult to use udev with a different init
system, but again most people don't care as long as the init system
they have works reliable. And since udev is Linux-only anyway, I don't
see a problem merging it with a Linux-only init system.

If it's so important to be able to choose such a low-level component
as the init system, why aren't people demanding that you can choose
different kernel stacks of choice? For example OSS4 instead of ALSA or
the old Firewire stack instead of the new one?

Adrian

--
.''`. John Paul Adrian Glaubitz
: :' : Debian Developer - glau...@debian.org
`. `' Freie Universitaet Berlin - glau...@physik.fu-berlin.de
`- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/20121125150...@physik.fu-berlin.de

[Henrique de Moraes Holschuh]

On Sun, 25 Nov 2012, John Paul Adrian Glaubitz wrote:
> On Sun, Nov 25, 2012 at 10:48:27PM +1300, Chris Bannister wrote:
> > https://lists.linuxaudio.org/mailarchive/lau/2012/11/21/194431
> >
> > There is a rather bad smell regarding all this.
>
> None of the systemd advocates ever mentioned for example the real
> reason why it uses such an ugly configuration syntax and Windoze
> format files. ^^^^^^^

The:

[crap]
foo = bar

format for config files is widely despised. And this is not a systemd
issue, even git uses that crap instead of something better like xml,
or simpler, like the hierarchical format used by apt that resembles
C++ classes.

--
"One disk to rule them all, One disk to find them. One disk to bring
them all and in the darkness grind them. In the Land of Redmond
where the shadows lie." -- The Silicon Valley Tarot
Henrique Holschuh

--
To UNSUBSCRIBE, email to debian-dev...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org

Archive: http://lists.debian.org/2012112515...@khazad-dum.debian.net