Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.os.linux.misc > #32816 > unrolled thread
| Started by | F Russell <fr@random.info> |
|---|---|
| First post | 2021-08-23 23:04 +0000 |
| Last post | 2021-08-25 00:12 -0400 |
| Articles | 20 on this page of 108 — 21 participants |
Back to article view | Back to comp.os.linux.misc
This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by
below is the oldest one visible, not the original post.
Re: What's the best linux distro? F Russell <fr@random.info> - 2021-08-23 23:04 +0000
Re: What's the best linux distro? "Carlos E.R." <robin_listas@es.invalid> - 2021-08-24 03:13 +0200
Re: What's the best linux distro? The Natural Philosopher <tnp@invalid.invalid> - 2021-08-24 14:19 +0100
Re: What's the best linux distro? Jimmy Johnson <Jimmy@disposable.invalid> - 2021-08-25 20:47 -0700
Re: What's the best linux distro? Robert Riches <spamtrap42@jacob21819.net> - 2021-08-26 04:05 +0000
Re: What's the best linux distro? The Natural Philosopher <tnp@invalid.invalid> - 2021-08-26 09:48 +0100
Re: What's the best linux distro? The Natural Philosopher <tnp@invalid.invalid> - 2021-08-26 09:46 +0100
Re: What's the best linux distro? Rich <rich@example.invalid> - 2021-08-26 12:11 +0000
Re: What's the best linux distro? Jasen Betts <usenet@revmaps.no-ip.org> - 2021-08-26 13:21 +0000
Re: What's the best linux distro? The Natural Philosopher <tnp@invalid.invalid> - 2021-08-26 16:53 +0100
Re: What's the best linux distro? The Natural Philosopher <tnp@invalid.invalid> - 2021-08-26 16:51 +0100
Re: What's the best linux distro? F Russell <fr@random.info> - 2021-08-24 12:53 +0000
Re: What's the best linux distro? SixOverFive <hae274b.net> - 2021-08-25 00:41 -0400
Re: What's the best linux distro? The Natural Philosopher <tnp@invalid.invalid> - 2021-08-25 10:18 +0100
Re: What's the best linux distro? "Carlos E.R." <robin_listas@es.invalid> - 2021-08-25 11:56 +0200
Re: What's the best linux distro? Richard Kettlewell <invalid@invalid.invalid> - 2021-08-25 12:10 +0100
Re: X-post and follow-Up (Was: What's the best linux distro?) "Carlos E.R." <robin_listas@es.invalid> - 2021-08-26 11:33 +0200
Re: X-post and follow-Up Bud Frede <frede@mouse-potato.com> - 2021-08-29 06:52 -0400
Re: X-post and follow-Up "Carlos E. R." <robin_listas@es.invalid> - 2021-08-29 14:02 +0200
Re: What's the best linux distro? The Natural Philosopher <tnp@invalid.invalid> - 2021-08-25 13:10 +0100
Re: What's the best linux distro? "Carlos E.R." <robin_listas@es.invalid> - 2021-08-25 14:36 +0200
Re: What's the best linux distro? Dan Espen <dan1espen@gmail.com> - 2021-08-25 08:53 -0400
Re: What's the best linux distro? "Carlos E.R." <robin_listas@es.invalid> - 2021-08-25 15:13 +0200
Re: What's the best linux distro? The Natural Philosopher <tnp@invalid.invalid> - 2021-08-25 14:31 +0100
Re: What's the best linux distro? Dan Espen <dan1espen@gmail.com> - 2021-08-25 12:14 -0400
Re: What's the best linux distro? "Carlos E.R." <robin_listas@es.invalid> - 2021-08-25 20:13 +0200
Re: What's the best linux distro? Stéphane CARPENTIER <sc@fiat-linux.fr> - 2021-08-26 20:57 +0000
Re: What's the best linux distro? "Carlos E. R." <robin_listas@es.invalid> - 2021-08-27 01:38 +0200
Re: What's the best linux distro? The Natural Philosopher <tnp@invalid.invalid> - 2021-08-27 02:04 +0100
Re: What's the best linux distro? "Carlos E. R." <robin_listas@es.invalid> - 2021-08-27 09:29 +0200
Re: What's the best linux distro? Marc Haber <mh+usenetspam1118@zugschl.us> - 2021-08-27 17:01 +0200
Re: What's the best linux distro? Bud Frede <frede@mouse-potato.com> - 2021-08-29 07:26 -0400
Re: What's the best linux distro? "Carlos E. R." <robin_listas@es.invalid> - 2021-08-29 14:27 +0200
Re: What's the best linux distro? Stéphane CARPENTIER <sc@fiat-linux.fr> - 2021-08-26 20:53 +0000
Re: What's the best linux distro? The Natural Philosopher <tnp@invalid.invalid> - 2021-08-27 02:01 +0100
Re: What's the best linux distro? Bobbie Sellers <bliss@mouse-potato.com> - 2021-08-25 08:23 -0700
Re: What's the best linux distro? Dan Espen <dan1espen@gmail.com> - 2021-08-25 12:21 -0400
Re: What's the best linux distro? "Carlos E. R." <robin_listas@es.invalid> - 2021-08-27 01:54 +0200
Re: What's the best linux distro? Stéphane CARPENTIER <sc@fiat-linux.fr> - 2021-08-28 12:51 +0000
Re: What's the best linux distro? Dan Espen <dan1espen@gmail.com> - 2021-08-28 11:40 -0400
Re: What's the best linux distro? Stéphane CARPENTIER <sc@fiat-linux.fr> - 2021-08-28 17:39 +0000
Re: What's the best linux distro? Dan Espen <dan1espen@gmail.com> - 2021-08-28 14:25 -0400
Re: What's the best linux distro? "Carlos E. R." <robin_listas@es.invalid> - 2021-08-28 23:08 +0200
Re: What's the best linux distro? Aragorn <thorongil@telenet.be> - 2021-08-29 00:10 +0200
Re: What's the best linux distro? Bud Frede <frede@mouse-potato.com> - 2021-08-29 08:09 -0400
Re: What's the best linux distro? "Carlos E. R." <robin_listas@es.invalid> - 2021-08-29 14:43 +0200
Re: What's the best linux distro? Bud Frede <frede@mouse-potato.com> - 2021-08-29 07:11 -0400
Re: What's the best linux distro? "Carlos E. R." <robin_listas@es.invalid> - 2021-08-29 14:29 +0200
Re: What's the best linux distro? Aragorn <thorongil@telenet.be> - 2021-08-29 16:09 +0200
Re: What's the best linux distro? The Natural Philosopher <tnp@invalid.invalid> - 2021-08-25 14:27 +0100
Re: What's the best linux distro? Robert Latest <boblatest@yahoo.com> - 2021-08-25 15:38 +0000
Re: What's the best linux distro? The Natural Philosopher <tnp@invalid.invalid> - 2021-08-25 18:44 +0100
Re: What's the best linux distro? "Carlos E.R." <robin_listas@es.invalid> - 2021-08-25 20:15 +0200
Re: What's the best linux distro? Robert Latest <boblatest@yahoo.com> - 2021-08-26 05:19 +0000
Re: What's the best linux distro? "Carlos E.R." <robin_listas@es.invalid> - 2021-08-26 11:40 +0200
Re: What's the best linux distro? Dan Espen <dan1espen@gmail.com> - 2021-08-25 12:07 -0400
Re: What's the best linux distro? Dan Espen <dan1espen@gmail.com> - 2021-08-25 12:10 -0400
Re: What's the best linux distro? Stéphane CARPENTIER <sc@fiat-linux.fr> - 2021-08-26 21:03 +0000
Re: What's the best linux distro? Marc Haber <mh+usenetspam1118@zugschl.us> - 2021-08-27 17:04 +0200
Re: What's the best linux distro? Stéphane CARPENTIER <sc@fiat-linux.fr> - 2021-08-28 13:12 +0000
Re: What's the best linux distro? Dan Espen <dan1espen@gmail.com> - 2021-08-28 11:39 -0400
Re: What's the best linux distro? Richard Kettlewell <invalid@invalid.invalid> - 2021-08-29 11:05 +0100
Re: What's the best linux distro? Dan Espen <dan1espen@gmail.com> - 2021-08-29 07:06 -0400
Re: What's the best linux distro? Richard Kettlewell <invalid@invalid.invalid> - 2021-08-29 12:50 +0100
Re: What's the best linux distro? Dan Espen <dan1espen@gmail.com> - 2021-08-29 08:07 -0400
Re: What's the best linux distro? Richard Kettlewell <invalid@invalid.invalid> - 2021-08-29 13:23 +0100
Re: What's the best linux distro? Dan Espen <dan1espen@gmail.com> - 2021-08-29 11:38 -0400
Re: What's the best linux distro? Richard Kettlewell <invalid@invalid.invalid> - 2021-08-29 22:45 +0100
Re: What's the best linux distro? "Carlos E. R." <robin_listas@es.invalid> - 2021-08-30 01:10 +0200
Re: What's the best linux distro? Dan Espen <dan1espen@gmail.com> - 2021-08-29 20:18 -0400
Re: What's the best linux distro? Richard Kettlewell <invalid@invalid.invalid> - 2021-08-30 08:23 +0100
Re: What's the best linux distro? Dan Espen <dan1espen@gmail.com> - 2021-08-30 08:02 -0400
Re: What's the best linux distro? The Natural Philosopher <tnp@invalid.invalid> - 2021-08-30 14:26 +0100
Re: What's the best linux distro? Richard Kettlewell <invalid@invalid.invalid> - 2021-08-31 10:13 +0100
Re: What's the best linux distro? The Natural Philosopher <tnp@invalid.invalid> - 2021-08-31 10:38 +0100
Re: What's the best linux distro? Eli the Bearded <*@eli.users.panix.com> - 2021-08-31 20:39 +0000
Re: What's the best linux distro? Stéphane CARPENTIER <sc@fiat-linux.fr> - 2021-08-29 12:59 +0000
Re: What's the best linux distro? "Carlos E. R." <robin_listas@es.invalid> - 2021-08-29 15:43 +0200
Re: What's the best linux distro? Dan Espen <dan1espen@gmail.com> - 2021-08-29 11:41 -0400
Re: What's the best linux distro? Richard Kettlewell <invalid@invalid.invalid> - 2021-08-26 09:09 +0100
Re: What's the best linux distro? The Natural Philosopher <tnp@invalid.invalid> - 2021-08-26 09:49 +0100
Re: What's the best linux distro? "Carlos E.R." <robin_listas@es.invalid> - 2021-08-26 11:31 +0200
Re: What's the best linux distro? Richard Kettlewell <invalid@invalid.invalid> - 2021-08-25 16:47 +0100
Re: What's the best linux distro? jjb <invalid@invalid.nl> - 2021-08-25 18:17 +0200
Re: What's the best linux distro? Bobbie Sellers <bliss@mouse-potato.com> - 2021-08-25 10:39 -0700
Re: What's the best linux distro? Aragorn <thorongil@telenet.be> - 2021-08-25 22:14 +0200
Re: What's the best linux distro? Bud Frede <frede@mouse-potato.com> - 2021-08-29 08:19 -0400
Re: What's the best linux distro? Dan Espen <dan1espen@gmail.com> - 2021-08-25 12:23 -0400
Re: What's the best linux distro? Richard Kettlewell <invalid@invalid.invalid> - 2021-08-25 18:14 +0100
Re: What's the best linux distro? "Carlos E.R." <robin_listas@es.invalid> - 2021-08-25 20:19 +0200
Re: What's the best linux distro? Bud Frede <frede@mouse-potato.com> - 2021-08-29 07:32 -0400
Re: What's the best linux distro? "Carlos E. R." <robin_listas@es.invalid> - 2021-08-29 14:36 +0200
Re: What's the best linux distro? The Natural Philosopher <tnp@invalid.invalid> - 2021-08-29 15:16 +0100
Re: What's the best linux distro? Jimmy Johnson <Jimmy@disposable.invalid> - 2021-08-25 21:02 -0700
Re: What's the best linux distro? Robert Latest <boblatest@yahoo.com> - 2021-08-26 05:35 +0000
Re: What's the best linux distro? "Carlos E.R." <robin_listas@es.invalid> - 2021-08-26 11:28 +0200
Re: What's the best linux distro? Aragorn <thorongil@telenet.be> - 2021-08-26 18:10 +0200
Re: What's the best linux distro? Dan Espen <dan1espen@gmail.com> - 2021-08-26 13:12 -0400
Re: What's the best linux distro? Aragorn <thorongil@telenet.be> - 2021-08-26 19:41 +0200
Re: What's the best linux distro? "J.O. Aho" <user@example.net> - 2021-08-27 11:02 +0200
Re: What's the best linux distro? Bud Frede <frede@mouse-potato.com> - 2021-08-29 07:42 -0400
Re: What's the best linux distro? "J.O. Aho" <user@example.net> - 2021-08-29 15:26 +0200
Re: What's the best linux distro? Richard Kettlewell <invalid@invalid.invalid> - 2021-08-29 14:57 +0100
Re: What's the best linux distro? Bud Frede <frede@mouse-potato.com> - 2021-08-29 10:47 -0400
Re: What's the best linux distro? Stéphane CARPENTIER <sc@fiat-linux.fr> - 2021-08-26 21:12 +0000
Re: What's the best linux distro? Branimir Maksimovic <branimir.maksimovic@gmail.com> - 2021-08-31 21:36 +0000
Re: What's the best linux distro? Stéphane CARPENTIER <sc@fiat-linux.fr> - 2021-08-26 20:41 +0000
Re: What's the best linux distro? SixOverFive <hae274b.net> - 2021-08-25 00:12 -0400
Page 5 of 6 — ← Prev page 1 2 3 4 [5] 6 Next page →
| From | The Natural Philosopher <tnp@invalid.invalid> |
|---|---|
| Date | 2021-08-26 09:49 +0100 |
| Message-ID | <sg7kis$8ta$5@dont-email.me> |
| In reply to | #32859 |
On 26/08/2021 09:09, Richard Kettlewell wrote: > The Natural Philosopher <tnp@invalid.invalid> writes: >> Having scripts execute in an orderly fashion according to >> alphanumerical order was perfectly OK >> >> You looked at what systems your layer depended on, amd made sure you >> were later in the list. > > If you’re rolling your own, that’s fine, although doing it by hand > rather than getting a computer to work it out for you seems rather > strange. > > If you’re building a distribution, with a wide range of possible end > user configurations, it’s not sustainable; you need something that will > work it out dynamically, based on the local configuration. IIRC Debian > had long standing bugs in this area until adopting systemd. > And not a few due to adopting systemd... -- The biggest threat to humanity comes from socialism, which has utterly diverted our attention away from what really matters to our existential survival, to indulging in navel gazing and faux moral investigations into what the world ought to be, whilst we fail utterly to deal with what it actually is.
[toc] | [prev] | [next] | [standalone]
| From | "Carlos E.R." <robin_listas@es.invalid> |
|---|---|
| Date | 2021-08-26 11:31 +0200 |
| Message-ID | <tj3kvh-5jq.ln1@Telcontar.valinor> |
| In reply to | #32862 |
On 26/08/2021 10.49, The Natural Philosopher wrote: > On 26/08/2021 09:09, Richard Kettlewell wrote: >> The Natural Philosopher <tnp@invalid.invalid> writes: >>> Having scripts execute in an orderly fashion according to >>> alphanumerical order was perfectly OK >>> >>> You looked at what systems your layer depended on, amd made sure you >>> were later in the list. >> >> If you’re rolling your own, that’s fine, although doing it by hand >> rather than getting a computer to work it out for you seems rather >> strange. >> >> If you’re building a distribution, with a wide range of possible end >> user configurations, it’s not sustainable; you need something that will >> work it out dynamically, based on the local configuration. IIRC Debian >> had long standing bugs in this area until adopting systemd. >> > And not a few due to adopting systemd... Which are being worked on or at least have hopes, not as the previous system. -- Cheers, Carlos.
[toc] | [prev] | [next] | [standalone]
| From | Richard Kettlewell <invalid@invalid.invalid> |
|---|---|
| Date | 2021-08-25 16:47 +0100 |
| Message-ID | <87zgt5jzkh.fsf@LkoBDZeT.terraraq.uk> |
| In reply to | #32826 |
"Carlos E.R." <robin_listas@es.invalid> writes: > The Natural Philosopher wrote: >> On 25/08/2021 10:56, Carlos E.R. wrote: >>> Systemd is evil. >> >> Well on that I do agree. >> >> Goes completely against the principles of simplicity, transparency, >> and not doing stuff that was working perfectly well already > > Well, the later is arguable. Many were not happy, and initd had no > maintenance or hopes of improvements, so they were looking for a > replacement. It could have been systemd or something else, but there > was nothing else suitable. So many distributions changed. “Nothing else suitable” isn’t true at all. Disregarding the rather large set of toy/experimental inits, Upstart was already deployed in production Linux systems. Launchd would have needed porting for use on Linux but was (and still is) happily managing large numbers of real computers. systemd as it is today has a lot of features they’re missing, certainly, but applying the same effort to them that was in fact put into systemd doesn’t seem like a totally unrealistic option, had anyone chosen to pursue it. -- https://www.greenend.org.uk/rjk/
[toc] | [prev] | [next] | [standalone]
| From | jjb <invalid@invalid.nl> |
|---|---|
| Date | 2021-08-25 18:17 +0200 |
| Message-ID | <sg5qe2$mtm$1@gioia.aioe.org> |
| In reply to | #32833 |
On 25-08-2021 17:47, Richard Kettlewell wrote: > systemd as it is today has a lot of features they’re missing, certainly, > but applying the same effort to them that was in fact put into systemd > doesn’t seem like a totally unrealistic option, had anyone chosen to > pursue it. > But they didn't...
[toc] | [prev] | [next] | [standalone]
| From | Bobbie Sellers <bliss@mouse-potato.com> |
|---|---|
| Date | 2021-08-25 10:39 -0700 |
| Message-ID | <sg5v8v$an9$1@dont-email.me> |
| In reply to | #32837 |
On 8/25/21 09:17, jjb wrote:
> On 25-08-2021 17:47, Richard Kettlewell wrote:
>> systemd as it is today has a lot of features they’re missing, certainly,
>> but applying the same effort to them that was in fact put into systemd
>> doesn’t seem like a totally unrealistic option, had anyone chosen to
>> pursue it.
>>
>
> But they didn't...
>
Because they see the featuritis of systemd as a complication
to be avoided rather than sought. One program one function
is the old GNU design criteria.
bliss - boots & runs a Pretty Cool Linux Operating System aka pclinuxos.
No to systemd.
[toc] | [prev] | [next] | [standalone]
| From | Aragorn <thorongil@telenet.be> |
|---|---|
| Date | 2021-08-25 22:14 +0200 |
| Message-ID | <20210825221425.1a2fc742@nx-74205> |
| In reply to | #32841 |
On 25.08.2021 at 10:39, Bobbie Sellers scribbled:
> On 8/25/21 09:17, jjb wrote:
>
> > On 25-08-2021 17:47, Richard Kettlewell wrote:
> >
> >> systemd as it is today has a lot of features they’re missing,
> >> certainly, but applying the same effort to them that was in fact
> >> put into systemd doesn’t seem like a totally unrealistic option,
> >> had anyone chosen to pursue it.
> >
> > But they didn't...
>
> Because they see the featuritis of systemd as a complication
> to be avoided rather than sought.
Among other things that have already been addressed and that I shan't
get into anymore, the kernel was offering many advanced possibilities to
userspace that userspace wasn't making use of -- cgroups. And to the
best of my knowledge, even today, only one other init system beside
systemd offers support for that, i.e. Gentoo's openrc.
I won't contradict that there have been some ego issues involved --
e.g. Lennart asking the GNOME developers to make systemd's logind a
hard dependency of GDM -- but systemd really IS progress. It makes for
a consistent and transparent yet modular system daemon that replaces a
poorly maintained and archaic init system that blindly started services
with virtually no control over restarting them or managing them once
they were running, leaving disparate ways of managing them.
> One program one function is the old GNU design criteria.
Actually, that goes back to UNIX in general, and you might be sad to
hear that the GNU find utility sins badly against that principle, not
to mention GNU emacs.
> bliss - boots & runs a Pretty Cool Linux Operating System aka
> pclinuxos.
>
> No to systemd.
I've been running Manjaro -- which is also cool, and very clean under
the hood -- since late April 2019. Manjaro is based upon Arch and uses
systemd. The three main differences with Arch proper are...
1. It has a graphical installer.
2. It has many graphical utilities for installiong software and
configuring the system, including a Manjaro-specific package
manager that can be used as an alternative to pacman while still
maintaining full compatibility with it, and that can be used
graphically or from the command line.
3. It's a curated rolling release, so instead of there being updates
every day, updates are tested, bundled together and pushed out on
average twice a month, barring urgent updates to certain packages
for stability or security reasons.
My computer boots up in 10 seconds from pressing Enter at the GRUB menu
to the SDDM login screen, and from a successful login to a fully
loaded Plasma desktop takes about 2 seconds longer.
Given that kernels are updated with new patches without forcing one to
adopt a higher kernel version, it is necessary to reboot the machine at
least twice a month, or whenever a major update is rolled out. Updates
are always announced on dedicated threads at the forum, together with a
list of possible gotchas and how to deal with them.
(But of course, the n00bs won't look at those announcements. They're
even too lazy to do a forum search on whatever issue they may be
encountering with an update, and then we as moderators have to merge
their threads and point them at the announcement thread for a relief
from their oh-so-tragic burdens. <rolling eyes>)
--
With respect,
= Aragorn =
[toc] | [prev] | [next] | [standalone]
| From | Bud Frede <frede@mouse-potato.com> |
|---|---|
| Date | 2021-08-29 08:19 -0400 |
| Message-ID | <IXKWI.15071$F26.9129@fx44.iad> |
| In reply to | #32841 |
Bobbie Sellers <bliss@mouse-potato.com> writes: > On 8/25/21 09:17, jjb wrote: >> On 25-08-2021 17:47, Richard Kettlewell wrote: >>> systemd as it is today has a lot of features they’re missing, certainly, >>> but applying the same effort to them that was in fact put into systemd >>> doesn’t seem like a totally unrealistic option, had anyone chosen to >>> pursue it. >>> >> But they didn't... >> > > Because they see the featuritis of systemd as a complication > to be avoided rather than sought. One program one function > is the old GNU design criteria. Emacs anyone? :-) I'd also point to compression being added to tar with GNU tar. That compbined two programs into one. Not very "one program one function." There are plenty of non-GNU examples of UNIX things that don't adhere to "one program one function" as well.
[toc] | [prev] | [next] | [standalone]
| From | Dan Espen <dan1espen@gmail.com> |
|---|---|
| Date | 2021-08-25 12:23 -0400 |
| Message-ID | <sg5qq2$2vm$5@dont-email.me> |
| In reply to | #32833 |
Richard Kettlewell <invalid@invalid.invalid> writes: > "Carlos E.R." <robin_listas@es.invalid> writes: >> The Natural Philosopher wrote: >>> On 25/08/2021 10:56, Carlos E.R. wrote: >>>> Systemd is evil. >>> >>> Well on that I do agree. >>> >>> Goes completely against the principles of simplicity, transparency, >>> and not doing stuff that was working perfectly well already >> >> Well, the later is arguable. Many were not happy, and initd had no >> maintenance or hopes of improvements, so they were looking for a >> replacement. It could have been systemd or something else, but there >> was nothing else suitable. So many distributions changed. > > “Nothing else suitable” isn’t true at all. Disregarding the rather large > set of toy/experimental inits, Upstart was already deployed in > production Linux systems. Launchd would have needed porting for use on > Linux but was (and still is) happily managing large numbers of real > computers. > > systemd as it is today has a lot of features they’re missing, certainly, > but applying the same effort to them that was in fact put into systemd > doesn’t seem like a totally unrealistic option, had anyone chosen to > pursue it. So now it's "some other init system could be better". That's no way to bash systemd, you're supposed to attack the character of the lead developer. -- Dan Espen
[toc] | [prev] | [next] | [standalone]
| From | Richard Kettlewell <invalid@invalid.invalid> |
|---|---|
| Date | 2021-08-25 18:14 +0100 |
| Message-ID | <87tujdjvkj.fsf@LkoBDZeT.terraraq.uk> |
| In reply to | #32839 |
Dan Espen <dan1espen@gmail.com> writes: > Richard Kettlewell <invalid@invalid.invalid> writes: >> "Carlos E.R." <robin_listas@es.invalid> writes: >>> The Natural Philosopher wrote: >>>> On 25/08/2021 10:56, Carlos E.R. wrote: >>>>> Systemd is evil. >>>> >>>> Well on that I do agree. >>>> >>>> Goes completely against the principles of simplicity, transparency, >>>> and not doing stuff that was working perfectly well already >>> >>> Well, the later is arguable. Many were not happy, and initd had no >>> maintenance or hopes of improvements, so they were looking for a >>> replacement. It could have been systemd or something else, but there >>> was nothing else suitable. So many distributions changed. >> >> “Nothing else suitable” isn’t true at all. Disregarding the rather large >> set of toy/experimental inits, Upstart was already deployed in >> production Linux systems. Launchd would have needed porting for use on >> Linux but was (and still is) happily managing large numbers of real >> computers. >> >> systemd as it is today has a lot of features they’re missing, certainly, >> but applying the same effort to them that was in fact put into systemd >> doesn’t seem like a totally unrealistic option, had anyone chosen to >> pursue it. > > So now it's "some other init system could be better". I’m not sure how you get to that from the above. Just pointing out that Carlos’s characterisation of the situation at the time decisions were being made wasn’t accurate. > That's no way to bash systemd, you're supposed to attack the character > of the lead developer. I’m not trying to bash anything here. -- https://www.greenend.org.uk/rjk/
[toc] | [prev] | [next] | [standalone]
| From | "Carlos E.R." <robin_listas@es.invalid> |
|---|---|
| Date | 2021-08-25 20:19 +0200 |
| Message-ID | <b7eivh-ga5.ln1@Telcontar.valinor> |
| In reply to | #32833 |
On 25/08/2021 17.47, Richard Kettlewell wrote: > "Carlos E.R." <robin_listas@es.invalid> writes: >> The Natural Philosopher wrote: >>> On 25/08/2021 10:56, Carlos E.R. wrote: >>>> Systemd is evil. >>> >>> Well on that I do agree. >>> >>> Goes completely against the principles of simplicity, transparency, >>> and not doing stuff that was working perfectly well already >> >> Well, the later is arguable. Many were not happy, and initd had no >> maintenance or hopes of improvements, so they were looking for a >> replacement. It could have been systemd or something else, but there >> was nothing else suitable. So many distributions changed. > > “Nothing else suitable” isn’t true at all. Disregarding the rather large > set of toy/experimental inits, Upstart was already deployed in > production Linux systems. Launchd would have needed porting for use on > Linux but was (and still is) happily managing large numbers of real > computers. Well, openSUSE was not about to use Ubuntu method :-p > systemd as it is today has a lot of features they’re missing, certainly, > but applying the same effort to them that was in fact put into systemd > doesn’t seem like a totally unrealistic option, had anyone chosen to > pursue it. > That's the point, no one chose to pursue it. -- Cheers, Carlos.
[toc] | [prev] | [next] | [standalone]
| From | Bud Frede <frede@mouse-potato.com> |
|---|---|
| Date | 2021-08-29 07:32 -0400 |
| Message-ID | <_eKWI.12682$rl3.1476@fx45.iad> |
| In reply to | #32846 |
"Carlos E.R." <robin_listas@es.invalid> writes: > On 25/08/2021 17.47, Richard Kettlewell wrote: >> "Carlos E.R." <robin_listas@es.invalid> writes: >>> The Natural Philosopher wrote: >>>> On 25/08/2021 10:56, Carlos E.R. wrote: >>>>> Systemd is evil. >>>> >>>> Well on that I do agree. >>>> >>>> Goes completely against the principles of simplicity, transparency, >>>> and not doing stuff that was working perfectly well already >>> >>> Well, the later is arguable. Many were not happy, and initd had no >>> maintenance or hopes of improvements, so they were looking for a >>> replacement. It could have been systemd or something else, but there >>> was nothing else suitable. So many distributions changed. >> “Nothing else suitable” isn’t true at all. Disregarding the rather >> large >> set of toy/experimental inits, Upstart was already deployed in >> production Linux systems. Launchd would have needed porting for use on >> Linux but was (and still is) happily managing large numbers of real >> computers. > > Well, openSUSE was not about to use Ubuntu method :-p > Redhat was using upstart. That's arguably more persuasive than anything that SuSE was doing or not doing. :-)
[toc] | [prev] | [next] | [standalone]
| From | "Carlos E. R." <robin_listas@es.invalid> |
|---|---|
| Date | 2021-08-29 14:36 +0200 |
| Message-ID | <kibsvhxcdt.ln2@minas-tirith.valinor> |
| In reply to | #32965 |
On 29/08/2021 13.32, Bud Frede wrote:
> "Carlos E.R." <robin_listas@es.invalid> writes:
>> On 25/08/2021 17.47, Richard Kettlewell wrote:
>>> "Carlos E.R." <robin_listas@es.invalid> writes:
>>>> The Natural Philosopher wrote:
>>>>> On 25/08/2021 10:56, Carlos E.R. wrote:
>>>>>> Systemd is evil.
>>>>>
>>>>> Well on that I do agree.
>>>>>
>>>>> Goes completely against the principles of simplicity, transparency,
>>>>> and not doing stuff that was working perfectly well already
>>>>
>>>> Well, the later is arguable. Many were not happy, and initd had no
>>>> maintenance or hopes of improvements, so they were looking for a
>>>> replacement. It could have been systemd or something else, but there
>>>> was nothing else suitable. So many distributions changed.
>>> “Nothing else suitable” isn’t true at all. Disregarding the rather
>>> large
>>> set of toy/experimental inits, Upstart was already deployed in
>>> production Linux systems. Launchd would have needed porting for use on
>>> Linux but was (and still is) happily managing large numbers of real
>>> computers.
>>
>> Well, openSUSE was not about to use Ubuntu method :-p
>>
>
> Redhat was using upstart. That's arguably more persuasive than anything
> that SuSE was doing or not doing. :-)
I don't remember what was said at the time about upstart, but basically
it was "no" :-)
Me, at the time I preferred initd scripts. Easy to change /one/ script
that misbehaves. But of course, each script was different with its own
learning curve. I did not see the reason to change. But change came and
now I came to accept/like it.
Doesn't mean I /know/ systemd. When I have to do something I often have
to ask.
--
Cheers,
Carlos E.R.
[toc] | [prev] | [next] | [standalone]
| From | The Natural Philosopher <tnp@invalid.invalid> |
|---|---|
| Date | 2021-08-29 15:16 +0100 |
| Message-ID | <sgg4sa$7uq$1@dont-email.me> |
| In reply to | #32975 |
On 29/08/2021 13:36, Carlos E. R. wrote: > I did not see the reason to change. But change came and > now I came to accept/like it. > > Doesn't mean I/know/ systemd. When I have to do something I often have > to ask. Its like postscript and X windows. Designed to be able to do anything, but finally tamed by putting in shims to ensue that there is a way simpler API that only uses the useful 1% of the features... But they are standards and we are stuck with them -- "Women actually are capable of being far more than the feminists will let them."
[toc] | [prev] | [next] | [standalone]
| From | Jimmy Johnson <Jimmy@disposable.invalid> |
|---|---|
| Date | 2021-08-25 21:02 -0700 |
| Message-ID | <sg73ph$bra$1@gioia.aioe.org> |
| In reply to | #32823 |
On 08/25/2021 02:56 AM, Carlos E.R. wrote: > F Russell is one of the local trolls, a fanatic. Linux has to be his way > or it is evil. Systemd is evil. He likes a linux with basically no > services, no init system, tailored solely for him. What the > distributions do, a system that works as is for many people is evil. > Linux for the masses is evil, it has to be for hackers only and must be > difficult to use. Besides slowing a system down, spying and selling user info, what did systemd bring to the table? Systemd was a lie from the very beginning when they said over and over it was faster. Our $ystem is full of lie$. -- James, Son of John
[toc] | [prev] | [next] | [standalone]
| From | Robert Latest <boblatest@yahoo.com> |
|---|---|
| Date | 2021-08-26 05:35 +0000 |
| Message-ID | <ioong9FalfqU4@mid.individual.net> |
| In reply to | #32852 |
Jimmy Johnson wrote: > Besides slowing a system down, spying and selling user info, what did > systemd bring to the table? Please elaborate on the "spying" bit. Never heard of that.
[toc] | [prev] | [next] | [standalone]
| From | "Carlos E.R." <robin_listas@es.invalid> |
|---|---|
| Date | 2021-08-26 11:28 +0200 |
| Message-ID | <ff3kvh-5jq.ln1@Telcontar.valinor> |
| In reply to | #32858 |
On 26/08/2021 07.35, Robert Latest wrote: > Jimmy Johnson wrote: >> Besides slowing a system down, spying and selling user info, what did >> systemd bring to the table? > > Please elaborate on the "spying" bit. Never heard of that. Me neither. Prove it. The source code is available to any. It is just FUD. -- Cheers, Carlos.
[toc] | [prev] | [next] | [standalone]
| From | Aragorn <thorongil@telenet.be> |
|---|---|
| Date | 2021-08-26 18:10 +0200 |
| Message-ID | <20210826181054.3878f85e@nx-74205> |
| In reply to | #32852 |
On 25.08.2021 at 21:02, Jimmy Johnson scribbled: > On 08/25/2021 02:56 AM, Carlos E.R. wrote: > > > F Russell is one of the local trolls, a fanatic. Linux has to be > > his way or it is evil. Systemd is evil. He likes a linux with > > basically no services, no init system, tailored solely for him. > > What the distributions do, a system that works as is for many > > people is evil. Linux for the masses is evil, it has to be for > > hackers only and must be difficult to use. > > Besides slowing a system down, [... My PCLinuxOS system with SysVinit booted up from hitting Enter at the GRUB menu to a character-mode console login prompt in about 50 seconds. My Manjaro system with systemd boots up from hitting Enter at the GRUB menu to the SDDM login screen in 10 seconds. Now, I could be grossly wrong, but I am still firmly convinced that 10 seconds is less than 50 seconds. > ...] spying and selling user info [... Links, please? To verifiable sources, please -- not to some "I hate systemd" website. >...], what did systemd bring to the table? Uniformity and consistency in configuration, a failsafe mechanism for the starting and managing of daemons, faster boot times, isolation and containerization of processes, consistency in the naming convention of attached peripherals -- e.g. drives, network interfaces -- as well as autodetection of the EFI partition, the root partition and the swap partition without needing any entries in /etc/fstab on drives with a GUID partition table, et al. AND... unlike with SysVinit, the code is actively being maintained. -- With respect, = Aragorn =
[toc] | [prev] | [next] | [standalone]
| From | Dan Espen <dan1espen@gmail.com> |
|---|---|
| Date | 2021-08-26 13:12 -0400 |
| Message-ID | <sg8i28$hlk$1@dont-email.me> |
| In reply to | #32874 |
Aragorn <thorongil@telenet.be> writes: > On 25.08.2021 at 21:02, Jimmy Johnson scribbled: > >> On 08/25/2021 02:56 AM, Carlos E.R. wrote: >> >> > F Russell is one of the local trolls, a fanatic. Linux has to be >> > his way or it is evil. Systemd is evil. He likes a linux with >> > basically no services, no init system, tailored solely for him. >> > What the distributions do, a system that works as is for many >> > people is evil. Linux for the masses is evil, it has to be for >> > hackers only and must be difficult to use. >> >> Besides slowing a system down, [... > > My PCLinuxOS system with SysVinit booted up from hitting Enter at the > GRUB menu to a character-mode console login prompt in about 50 seconds. > > My Manjaro system with systemd boots up from hitting Enter at the GRUB > menu to the SDDM login screen in 10 seconds. > > Now, I could be grossly wrong, but I am still firmly convinced that 10 > seconds is less than 50 seconds. > >> ...] spying and selling user info [... > > Links, please? To verifiable sources, please -- not to some "I hate > systemd" website. > >>...], what did systemd bring to the table? > > Uniformity and consistency in configuration, a failsafe mechanism for > the starting and managing of daemons, faster boot times, isolation and > containerization of processes, consistency in the naming convention of > attached peripherals -- e.g. drives, network interfaces -- as well as > autodetection of the EFI partition, the root partition and the swap > partition without needing any entries in /etc/fstab on drives with a > GUID partition table, et al. > > AND... unlike with SysVinit, the code is actively being maintained. Still have me on ignore? Good post, finally some reason on the subject. -- Dan Espen
[toc] | [prev] | [next] | [standalone]
| From | Aragorn <thorongil@telenet.be> |
|---|---|
| Date | 2021-08-26 19:41 +0200 |
| Message-ID | <20210826194125.32e586b0@nx-74205> |
| In reply to | #32884 |
On 26.08.2021 at 13:12, Dan Espen scribbled:
> Aragorn <thorongil@telenet.be> writes:
>
> > On 25.08.2021 at 21:02, Jimmy Johnson scribbled:
> >
> >> On 08/25/2021 02:56 AM, Carlos E.R. wrote:
> >>
> >> > F Russell is one of the local trolls, a fanatic. Linux has to be
> >> > his way or it is evil. Systemd is evil. He likes a linux with
> >> > basically no services, no init system, tailored solely for him.
> >> > What the distributions do, a system that works as is for many
> >> > people is evil. Linux for the masses is evil, it has to be for
> >> > hackers only and must be difficult to use.
> >>
> >> Besides slowing a system down, [...
> >
> > My PCLinuxOS system with SysVinit booted up from hitting Enter at
> > the GRUB menu to a character-mode console login prompt in about 50
> > seconds.
> >
> > My Manjaro system with systemd boots up from hitting Enter at the
> > GRUB menu to the SDDM login screen in 10 seconds.
> >
> > Now, I could be grossly wrong, but I am still firmly convinced that
> > 10 seconds is less than 50 seconds.
> >
> >> ...] spying and selling user info [...
> >
> > Links, please? To verifiable sources, please -- not to some "I hate
> > systemd" website.
> >
> >>...], what did systemd bring to the table?
> >
> > Uniformity and consistency in configuration, a failsafe mechanism
> > for the starting and managing of daemons, faster boot times,
> > isolation and containerization of processes, consistency in the
> > naming convention of attached peripherals -- e.g. drives, network
> > interfaces -- as well as autodetection of the EFI partition, the
> > root partition and the swap partition without needing any entries
> > in /etc/fstab on drives with a GUID partition table, et al.
> >
> > AND... unlike with SysVinit, the code is actively being maintained.
>
> Still have me on ignore?
Gosh, no -- that was a long time (and two desktop computers) ago.
> Good post, finally some reason on the subject.
I used to be anti-systemd too, because I couldn't see its merits, and I
guess I was just jerking my knee, like so many others. (I am only
human (I think) and not above making mistakes. ;))
But the more I looked into it, the more systemd simply started making
sense. It's progress. And now that I am running a distribution that
uses systemd [*], I don't see myself ever going back to SysVinit.
[*] Manjaro started off as systemd-based-only, given that it's a
derivative of Arch, but it has in the past also also supported
openrc. The openrc support was however dropped a number of years
ago, which is why a fork of Manjaro was created, called Artix.
--
With respect,
= Aragorn =
[toc] | [prev] | [next] | [standalone]
| From | "J.O. Aho" <user@example.net> |
|---|---|
| Date | 2021-08-27 11:02 +0200 |
| Message-ID | <ioro0fFso9eU1@mid.individual.net> |
| In reply to | #32887 |
On 26/08/2021 19.41, Aragorn wrote: > I used to be anti-systemd too, because I couldn't see its merits, and I > guess I was just jerking my knee, like so many others. (I am only > human (I think) and not above making mistakes. ;)) Sure systemd has come with some merits, but not everything has been great, as it do "eat up" other projects, which has lead to a number of serious security issues with a high exportability as it's been the pid 1 that been using them. Other init systems do handle the "benefit" that has been preached as the main reason to use systemd, the boot up time. For me from Grub to Kodi takes just 25 sec on OpenRC, would I want to monitor services, sure I could use s6 and still get the same bootup time. I do think that systemd has done a good job on possible service isolation, but it's doable on other init systems too, but requires a lot more hands on. For me, who has work quite many years with systems that had to have high security and prompt updates of high security vulnerabilities without too much impact on down time, systemd has been a bad option compared with say SysVinit. > [*] Manjaro started off as systemd-based-only, given that it's a > derivative of Arch, but it has in the past also also supported > openrc. The openrc support was however dropped a number of years > ago, which is why a fork of Manjaro was created, called Artix. Artix claims to be a based on Arch and it do use the Arch repositories and not the Manajro ones and you also notice that some packages are released faster on Artix than on Manjaro. -- //Aho
[toc] | [prev] | [next] | [standalone]
Page 5 of 6 — ← Prev page 1 2 3 4 [5] 6 Next page →
Back to top | Article view | comp.os.linux.misc
csiph-web