View previous topic :: View next topic |
Author |
Message |
The Doctor Moderator
Joined: 27 Jul 2010 Posts: 2678
|
Posted: Sat Jul 14, 2018 2:01 am Post subject: |
|
|
skorefish wrote: | Now we can be vigilant and report such dependencies.
When certain packages start pulling in systemd. Users like me can report them. Is there a kind of a list for doing so? | Unfortunately the only way to do that would be on the bug tracker for the individual project $upstream. Your mileage may vary depending on the mood of whom ever reads your report. _________________ First things first, but not necessarily in that order.
Apologies if I take a while to respond. I'm currently working on the dematerialization circuit for my blue box. |
|
Back to top |
|
|
skorefish Apprentice
Joined: 21 Jun 2015 Posts: 285
|
Posted: Sat Jul 14, 2018 5:38 am Post subject: |
|
|
Unfortunately indeed. I think, when projects are not responding, this path leads to forking or at least writing lots of patches, causing a hell lot of mayhem. Am i right?
So the more functions added to systemd, the more work for developers wanting to support "our freedom of choice" |
|
Back to top |
|
|
The Doctor Moderator
Joined: 27 Jul 2010 Posts: 2678
|
Posted: Sat Jul 14, 2018 6:02 am Post subject: |
|
|
That or use alternative software. In general, if a project or company is hostile to the user base I prefer to simply not give them my business at all. In Linux there is an alternative for almost everything. _________________ First things first, but not necessarily in that order.
Apologies if I take a while to respond. I'm currently working on the dematerialization circuit for my blue box. |
|
Back to top |
|
|
skorefish Apprentice
Joined: 21 Jun 2015 Posts: 285
|
Posted: Sat Jul 14, 2018 6:35 am Post subject: |
|
|
yep, i share your opinion, having alternatives makes linux strong |
|
Back to top |
|
|
steveL Watchman
Joined: 13 Sep 2006 Posts: 5153 Location: The Peanut Gallery
|
Posted: Sun Jul 15, 2018 12:08 pm Post subject: |
|
|
Naib wrote: | it is a shame they didn't do something about it instead of going "meh... so what" | skorefish wrote: | This is what i mean by defending our freedom of choice. | Yes, well they didn't, and they don't, so all the talk from Naib is simply hot-air, imo.
He is really stretching the definition of "freedom" to include the "freedom to abuse", or ignore it and pretend it has no impact on choice in general. skorefish wrote: | Now we can be vigilant and report such dependencies.
When certain packages start pulling in systemd. Users like me can report them. Is there a kind of a list for doing so? | Sorry, but this actually made me laugh. Gentoo devs generally hate being told anything at all; they usually only grudgingly accept lucid, comprehensive explanations of where they've gone wrong (a simple sentence or two just will not do) and then pretend like it never happened.
Most upstreams do not have a clue about packaging for distribution, so they quite happily accept the kool-aid of someone else doing that work.
As TheDoctor said, though, you will have a better chance with upstream; programmers usually want their code to work on more platforms rather than less. So if you get the patch upstreamed, then file the bug in Gentoo to get it version-bumped, you can sidestep the infamous Gentoo dev unpleasantness. |
|
Back to top |
|
|
1clue Advocate
Joined: 05 Feb 2006 Posts: 2569
|
Posted: Mon Jul 16, 2018 2:24 pm Post subject: |
|
|
Trying to say this without getting political about it. Moderators feel free to modify or remove this post if it crosses a line.
Vigilance: The same organization which sponsors systemd also sponsors many of the other packages with hard links to systemd. This is a for-profit organization with a lot of money, which has had designs on this sort of improvement for decades, and is clearly successful at least in part. Notifying them that one of the many packages they sponsor suddenly has a hard link to systemd will get a response similar to, "Nice that you noticed." Distro managers have little choice but to pull upstream changes. As the gnome-without-systemd group can attest, it gets increasingly difficult to remove the dependencies when the upstream developers are enthusiastically adopting it.
The raging political debate on this forum and in other places centers around whether these changes are good or not, and whether other packages sponsored by other organizations should have such a hard link. I'm not going to voice an opinion one way or the other here. I manage multiple systems with systemd and multiple systems without.
Freedom to abuse: The people making the changes consider them to be beneficial to Linux in general. It's undisputed AFAICT that those changes are financially beneficial to the organization sponsoring systemd.
Freedom of choice: There is a lot of choice in the software you use on Linux, especially concerning window managers and desktop software. Some of them follow the traditional windows-like gui desktop theme (e.g. kde, lxde) and some do something very different (e.g. i3, fvwm). Granted fvwm is antique software but it still works and still has a loyal following. To my (not-comprehensive) knowledge, none of those has hard to an init system links except gnome and that crowd. |
|
Back to top |
|
|
skorefish Apprentice
Joined: 21 Jun 2015 Posts: 285
|
Posted: Mon Jul 16, 2018 3:20 pm Post subject: |
|
|
Quote: | The same organization which sponsors systemd also sponsors many of the other packages with hard links to systemd. |
The first thing that struck me, trying systemd is they call daemons services. (like wicrosoft services)
So last time, i rode somewhere wicrosoft is sponsoring linux!!!? Is Wicrosoft sponsoring systemd behind the screens???!
about sponsors:
http://0pointer.net/blog/second-round-of-systemdconf-2015-sponsors.html |
|
Back to top |
|
|
1clue Advocate
Joined: 05 Feb 2006 Posts: 2569
|
Posted: Mon Jul 16, 2018 3:34 pm Post subject: |
|
|
skorefish wrote: | Quote: | The same organization which sponsors systemd also sponsors many of the other packages with hard links to systemd. |
The first thing that struck me, trying systemd is they call daemons services. (like wicrosoft services)
So last time, i rode somewhere wicrosoft is sponsoring linux!!!? Is Wicrosoft sponsoring systemd behind the screens???!
about sponsors:
http://0pointer.net/blog/second-round-of-systemdconf-2015-sponsors.html |
No. Systemd predates any interest Microsoft had in Linux by years. They "went public" in the 90s. They worked toward a commercially supported Linux and then started making changes which were sometimes met with enthusiasm and sometimes not.
I recommend a search engine and enter 'systemd', 'systemd debate', etc. Prepare to spend days reading relevant and conflicting information. Note that different sites support different views, and different authors as well. So the information you get from one site will not necessarily match information from another site.
Not surprisingly, sites and authors related to the organization which wrote systemd tend to show statistics indicating that systemd is a fantastic improvement in every way, and sites and authors opposed to it show very different test numbers. |
|
Back to top |
|
|
1clue Advocate
Joined: 05 Feb 2006 Posts: 2569
|
Posted: Mon Jul 16, 2018 3:39 pm Post subject: |
|
|
UNIX, from the time it was multitasking, has called the background tasks which provide services "daemons." I'm not sure exactly when that was, but I'm 52 and I believe the first daemons came into being before I could read or write. |
|
Back to top |
|
|
Anon-E-moose Watchman
Joined: 23 May 2008 Posts: 6147 Location: Dallas area
|
Posted: Mon Jul 16, 2018 3:52 pm Post subject: |
|
|
1clue wrote: | UNIX, from the time it was multitasking, has called the background tasks which provide services "daemons." I'm not sure exactly when that was, but I'm 52 and I believe the first daemons came into being before I could read or write. |
From wikipedia
The term was coined by the programmers of MIT's Project MAC. They took the name from Maxwell's demon, an imaginary being from a thought experiment that constantly works in the background, sorting molecules.[2] Unix systems inherited this terminology. Maxwell's Demon is consistent with Greek mythology's interpretation of a daemon as a supernatural being working in the background, with no particular bias towards good or evil. _________________ UM780, 6.1 zen kernel, gcc 13, profile 17.0 (custom bare multilib), openrc, wayland |
|
Back to top |
|
|
1clue Advocate
Joined: 05 Feb 2006 Posts: 2569
|
Posted: Mon Jul 16, 2018 4:09 pm Post subject: |
|
|
Anon-E-moose wrote: | 1clue wrote: | UNIX, from the time it was multitasking, has called the background tasks which provide services "daemons." I'm not sure exactly when that was, but I'm 52 and I believe the first daemons came into being before I could read or write. |
From wikipedia
The term was coined by the programmers of MIT's Project MAC. They took the name from Maxwell's demon, an imaginary being from a thought experiment that constantly works in the background, sorting molecules.[2] Unix systems inherited this terminology. Maxwell's Demon is consistent with Greek mythology's interpretation of a daemon as a supernatural being working in the background, with no particular bias towards good or evil. |
Interesting, and thanks. So the term predates my birth. And it also predates Microsoft. |
|
Back to top |
|
|
Naib Watchman
Joined: 21 May 2004 Posts: 6065 Location: Removed by Neddy
|
Posted: Mon Jul 16, 2018 4:16 pm Post subject: |
|
|
skorefish wrote: | Quote: | The same organization which sponsors systemd also sponsors many of the other packages with hard links to systemd. |
The first thing that struck me, trying systemd is they call daemons services. (like wicrosoft services)
So last time, i rode somewhere wicrosoft is sponsoring linux!!!? Is Wicrosoft sponsoring systemd behind the screens???!
about sponsors:
http://0pointer.net/blog/second-round-of-systemdconf-2015-sponsors.html | careful this is starting to feel like a generic systemd thread and thus will get merged into one of the other such threads.
That is why I made my post querying with regards to viewpoint. Its a system engineering concept where different views of the same system are valid in their own right to further the overall understanding. By posting as I did it facilitated drawing out the initial and valid viewpoint of ensuring unintentional tie-ins
If this thread starts taking the viewpoint of general systemd it will unfortunately become a candidate for merging. _________________
Quote: | Removed by Chiitoo |
|
|
Back to top |
|
|
The Doctor Moderator
Joined: 27 Jul 2010 Posts: 2678
|
Posted: Mon Jul 16, 2018 4:30 pm Post subject: |
|
|
Naib wrote: | If this thread starts taking the viewpoint of general systemd it will unfortunately become a candidate for merging. | Close it is.
Although as it started as a legit support request my inclination would be to lock it rather than merge it, but that might just be me. But please, do try to keep this on topic or take the discussion to the appropriate politics thread. I prefer moderating spammers, not the community. _________________ First things first, but not necessarily in that order.
Apologies if I take a while to respond. I'm currently working on the dematerialization circuit for my blue box. |
|
Back to top |
|
|
1clue Advocate
Joined: 05 Feb 2006 Posts: 2569
|
Posted: Mon Jul 16, 2018 4:33 pm Post subject: |
|
|
That might be my fault. I answered the questions because it was the OP asking them, and as I mentioned in that post I knew it was getting close to the line. |
|
Back to top |
|
|
skorefish Apprentice
Joined: 21 Jun 2015 Posts: 285
|
Posted: Mon Jul 16, 2018 5:53 pm Post subject: |
|
|
Ok, for me the above is part of the question.
(should)/Can we get rid of systemd ???
But please no more details about the politics because there are other threads about that.
Maybe one can post a link.
What i was thinking about. Do other init systems get sponsoring?
Last edited by skorefish on Mon Jul 16, 2018 6:09 pm; edited 2 times in total |
|
Back to top |
|
|
Anon-E-moose Watchman
Joined: 23 May 2008 Posts: 6147 Location: Dallas area
|
Posted: Mon Jul 16, 2018 6:06 pm Post subject: |
|
|
skorefish wrote: | What i was thinking about. Do other init systems get sponsoring? |
Basically RH is behind systemd, but AFAIK the rest of the init systems aren't backed by anyone but individual developers. If that's what you're wondering. _________________ UM780, 6.1 zen kernel, gcc 13, profile 17.0 (custom bare multilib), openrc, wayland |
|
Back to top |
|
|
skorefish Apprentice
Joined: 21 Jun 2015 Posts: 285
|
|
Back to top |
|
|
Anon-E-moose Watchman
Joined: 23 May 2008 Posts: 6147 Location: Dallas area
|
Posted: Mon Jul 16, 2018 6:38 pm Post subject: |
|
|
Two things I'll mention openrc isn't an init system, per se.
The other is Hubbs controls openrc and he's clearly in sys-d/RH's corner and seemingly wants to marginalize openrc in favor of sys-d.
At least that's my take based on his previous bonehead contributions to openrc.
Edit to add: No, I don't plan on going much further into sys-d territory as it's been covered in other places. _________________ UM780, 6.1 zen kernel, gcc 13, profile 17.0 (custom bare multilib), openrc, wayland |
|
Back to top |
|
|
Naib Watchman
Joined: 21 May 2004 Posts: 6065 Location: Removed by Neddy
|
Posted: Mon Jul 16, 2018 6:48 pm Post subject: |
|
|
Anon-E-moose wrote: | Two things I'll mention openrc isn't an init system, per se.
The other is Hubbs controls openrc and he's clearly in sys-d/RH's corner and seemingly wants to marginalize openrc in favor of sys-d.
At least that's my take based on his previous bonehead contributions to openrc.
Edit to add: No, I don't plan on going much further into sys-d territory as it's been covered in other places. | it is, since about 0.23 it has come with its own PID1
There have been three prolific criticism to counter openRC as a viable "modern init system"
1) it isn't PID1
2) no socket activation
3) no daemon keep-alive
Now no one has actually stated whether these are valid but are used to judge... but even then #1 and #3 are possible with OpenRC
Code: |
equery b openrc-init
* Searching for openrc-init ...
sys-apps/openrc-0.38.1 (/sbin/openrc-init)
|
https://forums.gentoo.org/viewtopic-t-1064510-highlight-openrcinit.html
https://wiki.gentoo.org/wiki/Comparison_of_init_systems _________________
Quote: | Removed by Chiitoo |
|
|
Back to top |
|
|
Naib Watchman
Joined: 21 May 2004 Posts: 6065 Location: Removed by Neddy
|
Posted: Mon Jul 16, 2018 7:09 pm Post subject: |
|
|
skorefish wrote: | Ok, for me the above is part of the question.
(should)/Can we get rid of systemd ???
But please no more details about the politics because there are other threads about that.
Maybe one can post a link.
What i was thinking about. Do other init systems get sponsoring? |
The Politics of systemd
The Politics of systemd Part 2
The Politics of systemd Part 3
but anyway...
how free of Systemd can a system be? well I do not have it and a large number of users also don't have it so it is possible. However... how many hard-depend on it and thus the end-user must make a choice between foo+systemd or !foo.
I am not sure how many ONLY depend on systemd and short of a crude grep over the tree I am not sure exactly how ... however as captured in this wiki ( https://wiki.gentoo.org/wiki/Gentoo_Without_systemd ) the list should only be associated with an init system (systemd or openrc) and not a specific package HOWEVER... gnome is one of those that pulled it in explicitly
--edit--
a crude grep has shown an annoying combination... mutter with the wayland USE flag
Code: | USE="wayland" emerge mutter -va
* IMPORTANT: 1 news items need reading for repository 'gentoo'.
* Use eselect news read to view new items.
These are the packages that would be merged, in order:
Calculating dependencies... done!
[ebuild N ] dev-libs/wayland-1.15.0::gentoo USE="-doc -static-libs" ABI_X86="32 (64) (-x32)" 423 KiB
[ebuild N ] dev-libs/wayland-protocols-1.14::gentoo 98 KiB
[ebuild N ] gnome-base/gnome-common-3.18.0-r1:3::gentoo USE="autoconf-archive" 153 KiB
[ebuild R ] media-libs/mesa-18.1.3::gentoo USE="classic dri3 egl gallium gbm llvm nptl wayland* -bindist -d3d9 -debug -gles1 -gles2 -opencl -openmax -osmesa -pax_kernel -pic (-selinux) -unwind -vaapi -valgrind -vdpau -vulkan -xa -xvmc" ABI_X86="32 (64) (-x32)" VIDEO_CARDS="(-freedreno) -i915 -i965 (-imx) -intel -nouveau -r100 -r200 -r300 -r600 -radeon -radeonsi (-vc4) -virgl (-vivante) -vmware" 0 KiB
[ebuild N ] sys-apps/systemd-239-r1:0/2::gentoo USE="acl gcrypt kmod lz4 pam pcre policykit resolvconf seccomp split-usr ssl sysv-utils -apparmor -audit -build -cryptsetup -curl -elfutils -gnuefi -http -idn -importd -libidn2 -lzma -nat -qrcode (-selinux) {-test} -vanilla -xkb" ABI_X86="32 (64) (-x32)" 7,004 KiB
[ebuild N ] sys-apps/gentoo-systemd-integration-7::gentoo 63 KiB
[ebuild R ] sys-apps/dbus-1.12.8::gentoo USE="X systemd* -debug -doc -elogind (-selinux) -static-libs {-test} -user-session" ABI_X86="32 (64) (-x32)" 0 KiB
[ebuild N ] sys-power/upower-0.99.7:0/3::gentoo USE="introspection -doc -ios (-selinux)" 438 KiB
[ebuild N ] gnome-base/gnome-desktop-3.24.2:3/12::gentoo USE="introspection udev -debug {-test}" 1,040 KiB
[ebuild R ] x11-base/xorg-server-1.20.0:0/1.20.0::gentoo USE="glamor ipv6 udev wayland* xorg xvfb -debug -dmx -doc -kdrive -libressl -minimal (-selinux) -static-libs -systemd -unwind -xcsecurity -xephyr -xnest" 5,954 KiB
[ebuild N ] gnome-extra/zenity-3.24.0::gentoo USE="libnotify webkit -debug" 1,060 KiB
[ebuild N ] x11-wm/mutter-3.24.4::gentoo USE="introspection udev wayland -debug -gles2 {-test}" INPUT_DEVICES="-wacom" 3,500 KiB
[blocks B ] sys-fs/eudev ("sys-fs/eudev" is blocking sys-apps/systemd-239-r1, sys-apps/gentoo-systemd-integration-7)
[blocks B ] sys-apps/systemd ("sys-apps/systemd" is blocking sys-fs/eudev-3.2.5)
[blocks B ] sys-apps/gentoo-systemd-integration ("sys-apps/gentoo-systemd-integration" is blocking sys-fs/eudev-3.2.5)
[blocks B ] sys-apps/sysvinit ("sys-apps/sysvinit" is blocking sys-apps/systemd-239-r1)
Total: 12 packages (9 new, 3 reinstalls), Size of downloads: 19,730 KiB
Conflict: 4 blocks (4 unsatisfied)
...
The following USE changes are necessary to proceed:
(see "package.use" in the portage(5) man page for more details)
# required by sys-apps/systemd-239-r1::gentoo
# required by sys-apps/gentoo-systemd-integration-7::gentoo
>=sys-apps/dbus-1.12.8 systemd
| so wayland depends on udev that is linked against systemd WHICH means eudev cannot be used ??? _________________
Quote: | Removed by Chiitoo |
|
|
Back to top |
|
|
Anon-E-moose Watchman
Joined: 23 May 2008 Posts: 6147 Location: Dallas area
|
Posted: Mon Jul 16, 2018 7:22 pm Post subject: |
|
|
Naib wrote: | Anon-E-moose wrote: | Two things I'll mention openrc isn't an init system, per se.
| it is, since about 0.23 it has come with its own PID1
|
AFAIK openrc depends on a real init under it, sysv, bsd, runit, etc and those indeed do run as PID 1.
I don't think you can run openrc without something under it.
It should be called an "init script provider." _________________ UM780, 6.1 zen kernel, gcc 13, profile 17.0 (custom bare multilib), openrc, wayland |
|
Back to top |
|
|
Naib Watchman
Joined: 21 May 2004 Posts: 6065 Location: Removed by Neddy
|
Posted: Mon Jul 16, 2018 7:23 pm Post subject: |
|
|
Anon-E-moose wrote: | Naib wrote: | Anon-E-moose wrote: | Two things I'll mention openrc isn't an init system, per se.
| it is, since about 0.23 it has come with its own PID1
|
AFAIK openrc depends on a real init under it, sysv, bsd, runit, etc and those indeed to run as PID 1.
I don't think you can run openrc without something under it.
It should be called an "init script provider." | nop openrc-init is pid1 if you set your system up to use it
I have provided the links and those either gave instructions OR also discussion about the source.
https://wiki.gentoo.org/wiki/OpenRC#openrc-init
Quote: |
Beginning with OpenRC 0.25, a new program is provided on Linux, openrc-init, which can replace /sbin/init on startup.
openrc-init will ignore the /etc/inittab file and boot OpenRC directly |
https://en.m.wikipedia.org/wiki/OpenRC
Quote: |
On Unix-like systems, OpenRC is a dependency-based init. Since 0.25 OpenRC includes openrc-init, which can replace /sbin/init, but the default provider for the init program is SysVinit for OpenRC |
My only error was stating 0.23, openrc-init came with its own pid1 in 0.25 _________________
Quote: | Removed by Chiitoo |
|
|
Back to top |
|
|
Anon-E-moose Watchman
Joined: 23 May 2008 Posts: 6147 Location: Dallas area
|
Posted: Mon Jul 16, 2018 7:42 pm Post subject: |
|
|
You seem to be correct on that.
Thank gawd I quit following openrc after the teen series.
openrc-init seems like another of Hubbs braindead ideas though. _________________ UM780, 6.1 zen kernel, gcc 13, profile 17.0 (custom bare multilib), openrc, wayland |
|
Back to top |
|
|
Naib Watchman
Joined: 21 May 2004 Posts: 6065 Location: Removed by Neddy
|
Posted: Mon Jul 16, 2018 7:48 pm Post subject: |
|
|
While some of hubb's actions have been... Debatable, this one seems valid, non-intrusive (you don't have to), code seems clean and does make OpenRC complete from an init system _________________
Quote: | Removed by Chiitoo |
|
|
Back to top |
|
|
Naib Watchman
Joined: 21 May 2004 Posts: 6065 Location: Removed by Neddy
|
Posted: Mon Jul 16, 2018 7:50 pm Post subject: |
|
|
Anyway...
So far there is GNOME and Wayland that do depend on systemd that are not optional .... Funtoo were able to provide a stub to permit gnome to function but Wayland? _________________
Quote: | Removed by Chiitoo |
|
|
Back to top |
|
|
|