Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.os.linux.misc > #11889 > unrolled thread
| Started by | noydb@no.way (Bruce Sinclair) |
|---|---|
| First post | 2014-09-02 23:35 +0000 |
| Last post | 2014-09-06 22:43 +0200 |
| Articles | 20 on this page of 181 — 36 participants |
Back to article view | Back to comp.os.linux.misc
What distros *don't* use systemd ? noydb@no.way (Bruce Sinclair) - 2014-09-02 23:35 +0000
Re: What distros *don't* use systemd ? Dan Espen <despen@verizon.net> - 2014-09-02 20:47 -0400
Re: What distros *don't* use systemd ? Andreas Kohlbach <sep14.5.ankman@spamgourmet.com> - 2014-09-03 18:20 -0400
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-03 23:40 +0100
Re: What distros *don't* use systemd ? "David W. Hodgins" <dwhodgins@nomail.afraid.org> - 2014-09-03 18:44 -0400
Re: What distros *don't* use systemd ? zeeb <zeeb@zeeb.zeeb> - 2014-09-10 23:16 -0500
Re: What distros *don't* use systemd ? Reinhardt Behm <rbehm@hushmail.com> - 2014-09-11 12:50 +0800
Re: What distros *don't* use systemd ? noydb@no.way (Bruce Sinclair) - 2014-09-07 23:06 +0000
Re: What distros *don't* use systemd ? Dan Espen <despen@verizon.net> - 2014-09-07 20:51 -0400
Re: What distros *don't* use systemd ? "Charles T. Smith" <cts.private.yahoo@gmail.com> - 2014-10-28 16:35 +0000
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-10-28 16:48 +0000
Re: What distros *don't* use systemd ? Dan Espen <despen@verizon.net> - 2014-10-28 14:18 -0400
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-10-28 22:00 +0000
Re: What distros *don't* use systemd ? Baho Utot <baho-utot@columbus.rr.com> - 2014-10-28 18:53 -0400
Re: What distros *don't* use systemd ? Aragorn <thorongil@telenet.be.invalid> - 2014-10-29 09:03 +0100
Re: What distros *don't* use systemd ? Baho Utot <baho-utot@columbus.rr.com> - 2014-10-29 17:32 -0400
Re: What distros *don't* use systemd ? "Charles T. Smith" <cts.private.yahoo@gmail.com> - 2014-11-01 09:17 +0000
Re: What distros *don't* use systemd ? Bobbie Sellers <bliss-sf4ever@dslextreme.com> - 2014-11-01 08:07 -0700
Re: What distros *don't* use systemd ? Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2014-11-02 04:58 +0000
Re: What distros *don't* use systemd ? Aragorn <thorongil@telenet.be.invalid> - 2014-11-02 06:59 +0100
Re: What distros *don't* use systemd ? Richard Kettlewell <rjk@greenend.org.uk> - 2014-11-02 09:57 +0000
Re: What distros *don't* use systemd ? Clark Smith <noaddress@nowhere.net> - 2014-11-02 14:40 +0000
Re: What distros *don't* use systemd ? Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2014-11-02 16:28 +0000
Re: What distros *don't* use systemd ? Bobbie Sellers <bliss-sf4ever@dslextreme.com> - 2014-11-02 07:27 -0800
Re: What distros *don't* use systemd ? Baho Utot <baho-utot@columbus.rr.com> - 2014-11-01 11:51 -0400
Re: What distros *don't* use systemd ? Rich <rich@example.invalid> - 2014-11-01 21:46 +0000
Re: What distros *don't* use systemd ? Baho Utot <baho-utot@columbus.rr.com> - 2014-11-01 18:24 -0400
Re: What distros *don't* use systemd ? Rich <rich@example.invalid> - 2014-11-01 15:30 +0000
Re: What distros *don't* use systemd ? Baho Utot <baho-utot@columbus.rr.com> - 2014-11-01 12:45 -0400
Re: What distros *don't* use systemd ? Michael Baeuerle <michael.baeuerle@gmx.net> - 2014-11-02 17:08 +0000
Re: What distros *don't* use systemd ? Baho Utot <baho-utot@columbus.rr.com> - 2014-11-02 15:59 -0500
Re: What distros *don't* use systemd ? William Unruh <unruh@invalid.ca> - 2014-11-02 22:22 +0000
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-11-01 20:13 +0000
Re: What distros *don't* use systemd ? Charlie Gibbs <cgibbs@kltpzyxm.invalid> - 2014-11-02 04:58 +0000
Re: What distros *don't* use systemd ? Dan Espen <despen@verizon.net> - 2014-10-28 14:15 -0400
Re: What distros *don't* use systemd ? Michael Baeuerle <michael.baeuerle@gmx.net> - 2014-10-28 21:02 +0000
Re: What distros *don't* use systemd ? "David W. Hodgins" <dwhodgins@nomail.afraid.org> - 2014-09-02 20:58 -0400
Re: What distros *don't* use systemd ? Robert Riches <spamtrap42@jacob21819.net> - 2014-09-03 03:12 +0000
Re: What distros *don't* use systemd ? noydb@no.way (Bruce Sinclair) - 2014-09-07 23:10 +0000
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-08 07:00 +0100
Re: What distros *don't* use systemd ? Dan Espen <despen@verizon.net> - 2014-09-08 10:51 -0400
Re: What distros *don't* use systemd ? notbob <notbob@nothome.com> - 2014-09-08 15:11 +0000
Re: What distros *don't* use systemd ? Aragorn <thorongil@telenet.be.invalid> - 2014-09-08 17:53 +0200
Re: What distros *don't* use systemd ? The Natural Philosopher <tnp@invalid.invalid> - 2014-09-08 17:26 +0100
Re: What distros *don't* use systemd ? Vilmos Soti <vilmos@soti.ca> - 2014-09-08 09:36 -0700
Re: What distros *don't* use systemd ? The Natural Philosopher <tnp@invalid.invalid> - 2014-09-08 17:42 +0100
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-08 17:52 +0100
Re: What distros *don't* use systemd ? The Natural Philosopher <tnp@invalid.invalid> - 2014-09-08 18:29 +0100
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-08 18:52 +0100
Re: Spam (thread drift) The Natural Philosopher <tnp@invalid.invalid> - 2014-09-08 18:58 +0100
Re: Spam (thread drift) Tim Watts <tw_usenet@dionic.net> - 2014-09-08 19:10 +0100
Re: Spam (thread drift) The Natural Philosopher <tnp@invalid.invalid> - 2014-09-08 19:18 +0100
Re: Spam (thread drift) Marc Haber <mh+usenetspam1118@zugschl.us> - 2014-09-08 22:01 +0200
Re: Spam (thread drift) Tim Watts <tw_usenet@dionic.net> - 2014-09-08 21:23 +0100
Re: Spam (thread drift) Marc Haber <mh+usenetspam1118@zugschl.us> - 2014-09-09 08:58 +0200
Re: Spam (thread drift) Tim Watts <tw_usenet@dionic.net> - 2014-09-09 09:58 +0100
Re: Spam (thread drift) Marc Haber <mh+usenetspam1118@zugschl.us> - 2014-09-09 11:12 +0200
Re: Spam (thread drift) Tim Watts <tw_usenet@dionic.net> - 2014-09-09 11:14 +0100
Re: Spam (thread drift) Tim Watts <tw_usenet@dionic.net> - 2014-09-09 11:16 +0100
Re: What distros *don't* use systemd ? Marc Haber <mh+usenetspam1118@zugschl.us> - 2014-09-08 21:59 +0200
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-08 17:30 +0100
Re: What distros *don't* use systemd ? notbob <notbob@nothome.com> - 2014-09-08 16:32 +0000
Re: What distros *don't* use systemd ? Dan Espen <despen@verizon.net> - 2014-09-08 13:31 -0400
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-08 17:28 +0100
Re: What distros *don't* use systemd ? The Natural Philosopher <tnp@invalid.invalid> - 2014-09-08 17:40 +0100
Re: What distros *don't* use systemd ? Dan Espen <despen@verizon.net> - 2014-09-08 13:41 -0400
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-08 19:05 +0100
Re: What distros *don't* use systemd ? Dan Espen <despen@verizon.net> - 2014-09-08 14:57 -0400
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-08 21:21 +0100
Re: What distros *don't* use systemd ? Dan Espen <despen@verizon.net> - 2014-09-08 17:59 -0400
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-09 07:25 +0100
Re: What distros *don't* use systemd ? John Hasler <jhasler@newsguy.com> - 2014-09-08 15:29 -0500
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-08 22:47 +0100
Re: What distros *don't* use systemd ? noydb@no.way (Bruce Sinclair) - 2014-09-08 23:20 +0000
Re: What distros *don't* use systemd ? John McCue <jmccue@jmclin0.bstnma.east.verizon.net> - 2014-09-09 00:37 +0000
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-09 07:27 +0100
Re: What distros *don't* use systemd ? Rich <rich@example.invalid> - 2014-09-09 10:19 +0000
Re: What distros *don't* use systemd ? Dan Espen <despen@verizon.net> - 2014-09-09 09:25 -0400
Re: What distros *don't* use systemd ? Balwinder S Dheeman <bdheeman.SANSPAM@outlook.com> - 2014-09-09 17:00 +0530
Re: What distros *don't* use systemd ? noydb@no.way (Bruce Sinclair) - 2014-09-09 23:11 +0000
Re: What distros *don't* use systemd ? The Natural Philosopher <tnp@invalid.invalid> - 2014-09-09 01:49 +0100
Re: What distros *don't* use systemd ? noydb@no.way (Bruce Sinclair) - 2014-09-09 04:03 +0000
Re: What distros *don't* use systemd ? William Poaster <wp@dev.null> - 2014-09-09 12:37 +0100
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-09 13:03 +0100
Re: What distros *don't* use systemd ? The Natural Philosopher <tnp@invalid.invalid> - 2014-09-09 13:13 +0100
Re: What distros *don't* use systemd ? Aragorn <thorongil@telenet.be.invalid> - 2014-09-09 14:49 +0200
Re: What distros *don't* use systemd ? noydb@no.way (Bruce Sinclair) - 2014-09-09 23:17 +0000
Re: What distros *don't* use systemd ? Rich <rich@example.invalid> - 2014-09-09 23:45 +0000
Re: What distros *don't* use systemd ? noydb@no.way (Bruce Sinclair) - 2014-09-09 23:14 +0000
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-09 07:29 +0100
Re: What distros *don't* use systemd ? John Hasler <jhasler@newsguy.com> - 2014-09-09 08:11 -0500
Re: What distros *don't* use systemd ? The Natural Philosopher <tnp@invalid.invalid> - 2014-09-09 15:03 +0100
Re: What distros *don't* use systemd ? notbob <notbob@nothome.com> - 2014-09-03 01:52 +0000
Re: What distros *don't* use systemd ? Rich <rich@example.invalid> - 2014-09-03 03:23 +0000
Re: What distros *don't* use systemd ? Keith Keller <kkeller-usenet@wombat.san-francisco.ca.us> - 2014-09-02 22:10 -0700
Re: What distros *don't* use systemd ? noydb@no.way (Bruce Sinclair) - 2014-09-07 23:12 +0000
Re: What distros *don't* use systemd ? Rich <rich@example.invalid> - 2014-09-08 00:51 +0000
Re: What distros *don't* use systemd ? Keith Keller <kkeller-usenet@wombat.san-francisco.ca.us> - 2014-09-07 18:50 -0700
Re: What distros *don't* use systemd ? Aragorn <thorongil@telenet.be.invalid> - 2014-09-08 08:00 +0200
Re: What distros *don't* use systemd ? Thad Floryan <thad@thadlabs.com> - 2014-09-07 23:08 -0700
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-08 10:09 +0100
Re: What distros *don't* use systemd ? The Natural Philosopher <tnp@invalid.invalid> - 2014-09-08 10:57 +0100
Re: What distros *don't* use systemd ? noydb@no.way (Bruce Sinclair) - 2014-09-08 03:57 +0000
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-08 07:02 +0100
Re: What distros *don't* use systemd ? Dan Espen <despen@verizon.net> - 2014-09-08 10:48 -0400
Re: What distros *don't* use systemd ? noydb@no.way (Bruce Sinclair) - 2014-09-08 23:32 +0000
Re: What distros *don't* use systemd ? John Hasler <jhasler@newsguy.com> - 2014-09-08 19:58 -0500
Re: What distros *don't* use systemd ? noydb@no.way (Bruce Sinclair) - 2014-09-09 04:04 +0000
Re: What distros *don't* use systemd ? "J@J.J" <j@j.j> - 2014-09-10 15:17 -0700
Re: What distros *don't* use systemd ? John McCue <jmccue@jmclin0.bstnma.east.verizon.net> - 2014-09-11 02:02 +0000
Re: What distros *don't* use systemd ? Keith Keller <kkeller-usenet@wombat.san-francisco.ca.us> - 2014-09-10 19:20 -0700
Re: What distros *don't* use systemd ? The Natural Philosopher <tnp@invalid.invalid> - 2014-09-11 08:27 +0100
Re: What distros *don't* use systemd ? Stef <not@this.address.com> - 2014-09-03 04:52 +0000
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-03 08:43 +0100
Re: What distros *don't* use systemd ? Stef <not@this.address.com> - 2014-09-03 14:40 +0000
Re: What distros *don't* use systemd ? Andreas Kohlbach <sep14.5.ankman@spamgourmet.com> - 2014-09-03 18:26 -0400
Re: What distros *don't* use systemd ? Marc Haber <mh+usenetspam1118@zugschl.us> - 2014-09-04 08:01 +0200
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-04 08:28 +0100
Re: What distros *don't* use systemd ? Marc Haber <mh+usenetspam1118@zugschl.us> - 2014-09-04 10:14 +0200
Re: What distros *don't* use systemd ? Aragorn <thorongil@telenet.be.invalid> - 2014-09-04 11:02 +0200
Re: What distros *don't* use systemd ? Bit Twister <BitTwister@mouse-potato.com> - 2014-09-04 09:33 +0000
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-04 11:16 +0100
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-04 11:13 +0100
Re: What distros *don't* use systemd ? Rich <rich@example.invalid> - 2014-09-04 11:03 +0000
Re: What distros *don't* use systemd ? Aragorn <thorongil@telenet.be.invalid> - 2014-09-04 14:22 +0200
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-04 15:06 +0100
Re: What distros *don't* use systemd ? gazelle@shell.xmission.com (Kenny McCormack) - 2014-09-04 12:26 +0000
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-04 15:09 +0100
Re: What distros *don't* use systemd ? Joe Beanfish <joebeanfish@nospam.duh> - 2014-09-05 13:15 +0000
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-05 15:16 +0100
Re: What distros *don't* use systemd ? Dan Espen <despen@verizon.net> - 2014-09-04 10:10 -0400
Re: What distros *don't* use systemd ? Aragorn <thorongil@telenet.be.invalid> - 2014-09-04 14:15 +0200
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-04 15:03 +0100
Re: What distros *don't* use systemd ? Benjamin Rochefort <ben@benwen.info> - 2014-09-04 20:23 -0400
Re: What distros *don't* use systemd ? Aragorn <thorongil@telenet.be.invalid> - 2014-09-05 06:41 +0200
Re: What distros *don't* use systemd ? Joe Beanfish <joebeanfish@nospam.duh> - 2014-09-05 13:23 +0000
Re: What distros *don't* use systemd ? William Unruh <unruh@invalid.ca> - 2014-09-05 13:47 +0000
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-05 15:26 +0100
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-05 15:17 +0100
Re: What distros *don't* use systemd ? William Unruh <unruh@invalid.ca> - 2014-09-05 20:51 +0000
Re: What distros *don't* use systemd ? The Natural Philosopher <tnp@invalid.invalid> - 2014-09-05 22:57 +0100
Re: What distros *don't* use systemd ? William Unruh <unruh@invalid.ca> - 2014-09-05 22:03 +0000
Re: What distros *don't* use systemd ? The Natural Philosopher <tnp@invalid.invalid> - 2014-09-05 23:25 +0100
Re: What distros *don't* use systemd ? William Unruh <unruh@invalid.ca> - 2014-09-06 01:52 +0000
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-06 07:49 +0100
Re: What distros *don't* use systemd ? Rich <rich@example.invalid> - 2014-09-06 13:11 +0000
Re: What distros *don't* use systemd ? notbob <notbob@nothome.com> - 2014-09-06 15:09 +0000
Re: What distros *don't* use systemd ? Keith Keller <kkeller-usenet@wombat.san-francisco.ca.us> - 2014-09-06 12:06 -0700
Re: What distros *don't* use systemd ? Rich <rich@example.invalid> - 2014-09-06 01:44 +0000
Re: What distros *don't* use systemd ? Rich <rich@example.invalid> - 2014-09-05 18:17 +0000
Re: What distros *don't* use systemd ? Joe Beanfish <joebeanfish@nospam.duh> - 2014-09-08 13:47 +0000
Re: What distros *don't* use systemd ? John Hasler <jhasler@newsguy.com> - 2014-09-08 09:15 -0500
Re: What distros *don't* use systemd ? Marc Haber <mh+usenetspam1118@zugschl.us> - 2014-09-04 12:24 +0200
Re: What distros *don't* use systemd ? Keith Keller <kkeller-usenet@wombat.san-francisco.ca.us> - 2014-09-04 07:53 -0700
Re: What distros *don't* use systemd ? "David W. Hodgins" <dwhodgins@nomail.afraid.org> - 2014-09-08 14:40 -0400
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-04 10:56 +0100
Re: What distros *don't* use systemd ? Marc Haber <mh+usenetspam1118@zugschl.us> - 2014-09-04 12:25 +0200
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-04 13:03 +0100
Re: What distros *don't* use systemd ? Marc Haber <mh+usenetspam1118@zugschl.us> - 2014-09-04 21:56 +0200
Re: What distros *don't* use systemd ? Aragorn <thorongil@telenet.be.invalid> - 2014-09-04 14:30 +0200
Re: What distros *don't* use systemd ? Stef <not@this.address.com> - 2014-09-04 15:42 +0000
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-04 17:19 +0100
Re: What distros *don't* use systemd ? Rich <rich@example.invalid> - 2014-09-04 22:35 +0000
Re: What distros *don't* use systemd ? Stef <not@this.address.com> - 2014-09-05 06:14 +0000
Re: What distros *don't* use systemd ? Balwinder S Dheeman <bdheeman.SANSPAM@outlook.com> - 2014-09-03 23:42 +0530
Re: What distros *don't* use systemd ? Rich <rich@example.invalid> - 2014-09-03 19:58 +0000
Re: What distros *don't* use systemd ? Jerry Peters <jerry@example.invalid> - 2014-09-03 20:36 +0000
Re: What distros *don't* use systemd ? Rich <rich@example.invalid> - 2014-09-03 21:40 +0000
Re: What distros *don't* use systemd ? Bobbie Sellers <bliss-sf4ever@dslextreme.com> - 2014-09-05 09:15 -0700
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-05 17:44 +0100
Re: What distros *don't* use systemd ? Keith Keller <kkeller-usenet@wombat.san-francisco.ca.us> - 2014-09-05 13:18 -0700
Re: What distros *don't* use systemd ? William Unruh <unruh@invalid.ca> - 2014-09-05 20:57 +0000
Re: What distros *don't* use systemd ? Rich <rich@example.invalid> - 2014-09-06 01:53 +0000
Re: What distros *don't* use systemd ? Aragorn <thorongil@telenet.be.invalid> - 2014-09-06 05:55 +0200
Re: What distros *don't* use systemd ? Tim Watts <tw_usenet@dionic.net> - 2014-09-06 07:53 +0100
Re: What distros *don't* use systemd ? Marc Haber <mh+usenetspam1118@zugschl.us> - 2014-09-06 10:49 +0200
Re: What distros *don't* use systemd ? Bobbie Sellers <bliss-sf4ever@dslextreme.com> - 2014-09-06 09:45 -0700
Re: What distros *don't* use systemd ? Keith Keller <kkeller-usenet@wombat.san-francisco.ca.us> - 2014-09-06 11:59 -0700
Re: What distros *don't* use systemd ? The Natural Philosopher <tnp@invalid.invalid> - 2014-09-06 21:29 +0100
Re: What distros *don't* use systemd ? Baho Uttot <baho-utot@columbus.rr.com> - 2014-09-07 07:04 -0400
Re: What distros *don't* use systemd ? Aragorn <thorongil@telenet.be.invalid> - 2014-09-06 22:43 +0200
Page 7 of 10 — ← Prev page 1 … 5 6 [7] 8 9 10 Next page →
| From | Bit Twister <BitTwister@mouse-potato.com> |
|---|---|
| Date | 2014-09-04 09:33 +0000 |
| Message-ID | <slrnm0gcg0.hmr.BitTwister@wb.home.test> |
| In reply to | #11923 |
On Thu, 04 Sep 2014 11:02:11 +0200, Aragorn wrote: > The reason why many distributions adopt systemd is because... > > systemd has brought a number of innovations and good ideas to GNU/Linux, > but the whole concept has been poorly implemented from the start, Fault path testing has also been lacking. > and > along with those good ideas and innovations also came a great number of > bad ideas, such as "the /usr move" [2], a 1 MiB monolith as PID 1, > system logs in a binary format, and very sloppy coding, to say the very > least. I hear that. > > In addition to all of this, the traditional System V init was never > broken. Embedded devices or things that need to be booted up multiple > times a day can benefit from the marginal increase in boot time offered > by systemd, but systems that remain up 24/7 don't care about boot times, > and especially not if those boot times are only marginally faster with > systemd than they are with the traditional System V init. Well, systemd boot up times have been getting better and better. You also have to keep in mind, those cloud servers and virtual guests need quicker boot times. :) I was getting impressed when I was getting less that 30 second boot times on my setup on earlier Mageia systemd releases. Testing/running Mageia 5 release as I type this. named is taking 25 seconds for shutdown, mysql daemon takes 5 minutes when systemd is trying to shutdown those stopped daemons. :-( Pretty stupid piece of code that has to time out stopping a stopped process because it did not respond to the shutdown request.
[toc] | [prev] | [next] | [standalone]
| From | Tim Watts <tw_usenet@dionic.net> |
|---|---|
| Date | 2014-09-04 11:16 +0100 |
| Message-ID | <9pqldb-3h6.ln1@squidward.local.dionic.net> |
| In reply to | #11924 |
On 04/09/14 10:33, Bit Twister wrote: > > You also have to keep in mind, those cloud servers and virtual guests > need quicker boot times. :) My "private cloud"[1] VMs boot debian in a matter of seconds with init.d/ [1] I hate all this "cloud" tagging - it's a virtual machine and/or a network service... On real hardware, the boto times of even Windows pales into insignificance compared to how long it takes a Dell PowerEdge[2] to even get to reading the boot sector of anything... [2] Otherwise great hardware IME - but why oh why does it take several minutes to boot! > I was getting impressed when I was getting less that 30 second boot > times on my setup on earlier Mageia systemd releases. > > Testing/running Mageia 5 release as I type this. named is taking 25 > seconds for shutdown, mysql daemon takes 5 minutes when systemd is > trying to shutdown those stopped daemons. :-( > > Pretty stupid piece of code that has to time out stopping a stopped > process because it did not respond to the shutdown request. >
[toc] | [prev] | [next] | [standalone]
| From | Tim Watts <tw_usenet@dionic.net> |
|---|---|
| Date | 2014-09-04 11:13 +0100 |
| Message-ID | <2kqldb-3h6.ln1@squidward.local.dionic.net> |
| In reply to | #11923 |
On 04/09/14 10:02, Aragorn wrote: > The reason why many distributions adopt systemd is because... > > a. They consider RedHat their upstream, and allow RedHat (and its > propaganda platform freedesktop.org) to dictate what they must > do. To the best of my knowledge, all .rpm-based distributions > have unfortunately stepped into this trap. Which doesn't explain debian (until you get to the Gnome bit further down). > b. systemd has been made into a hard dependency of GNOME/GDM as of > GNOME 3.8, which, like systemd, is developed over at RedHat [1]. > At the moment and barring more recent information, I believe that > this doesn't require systemd to be running as PID 1 yet, but > GNOME/GDM does now require systemd to be _installed_ either way. That should have been stopped at birth and is one of the reasons I use xfce4 (s/xfce4/your-fav-lightweight-win-manager/) Both Gnome and KDE have been buggers for innovating in the wrong place and putting features in the WM/DE that would have been far more sensibly made into a little clean daemon which could interact with any desktop (or none at all for terminal users) via a common messaging interface (which is dbus I guess). Taken to extremes, this approach would have had us only able to run KDE apps on a KDE DE and Gnome apps on Gnome. Silly example but it does follow as a logical extreme. Fortunately, the greater good prevailed there. > systemd has brought a number of innovations and good ideas to GNU/Linux, > but the whole concept has been poorly implemented from the start, and > along with those good ideas and innovations also came a great number of > bad ideas, such as > "the /usr move" [2], Oh god - I don't even know what that is... > a 1 MiB monolith as PID 1, That's idiotic and kills it for embedded systems. > system logs in a binary format, Do they think they're Microsoft? Even more idiot points for that. > and very sloppy coding, to say the very > least. And a poor attitude to bug reports as per my last post. > systemd violates the UNIX philosophy of having small, distinct and > interoperative tools that do one thing but do it excellently. There's a > good reason as to why UNIX has survived for over 40 years: it works and > it's easily maintainable. The arrogance displayed by Sievers and > Poettering [3] is a different matter altogether, but nevertheless > doesn't help in remedying the situation. For me, I will be putting it in the same can as Wayland - a cute idea at its core, but so far off the mark in practice I am not touching it with a bargepole. > In addition to all of this, the traditional System V init was never > broken. Embedded devices or things that need to be booted up multiple > times a day can benefit from the marginal increase in boot time offered > by systemd, but systems that remain up 24/7 don't care about boot times, > and especially not if those boot times are only marginally faster with > systemd than they are with the traditional System V init. Indeed. Full agreement. > > [1] It is obviously a strategic move by RedHat, who are trying to usurp > the market of enterprise-oriented GNU/Linux distributors, at the > cost of flexibility and choice towards the users and staying with > the proven UNIX tradition. The same attitude can be seen among the > GNOME and even the GTK developers. > > [2] While the current argument for "the /usr move" - i.e. that with all > binaries under /usr, it would be easier to have /usr mounted read- > only - makes sense, that was not the reason behind it in the first > place. The reason was that Poettering and Sievers had screwed up > and put boot-essential stuff under /usr - where it doesn't belong - > so as to placate a few corner cases. Foolish... There have been occasions when I have had to rescue something with a broken /usr/ (disk problem). > [3] "We are RedHat. Resistance is futile. You will be assimilated." > Lennart Poettering has more than once stated on his blog that he > joyfully looks forward to the day where no other init systems than > systemd will exist anymore, and has also equally arrogantly mocked > other distributions - among whom Gentoo - on his blog. > Redhat might still have a certain following[1], but in my world of linux in the UK I see a *lot* of debian and ubuntu deployment - both in academia and in startup tech companies in London. [1] Notably in banks and CERN/High energy physics as Scientific Linux is CERN's supported flavour and that's based off RHEL/Centos and originates from FermiLab The only thing Centos/RHEL is good for is its hugely long support lifetimes of certain releases. But I persoanally find it very frustrating to work with compared to Debian Stable. In fact I've even gone to Debian on my desktop (but Mint on my laptop). It takes a little more effort to polish it up but not much...
[toc] | [prev] | [next] | [standalone]
| From | Rich <rich@example.invalid> |
|---|---|
| Date | 2014-09-04 11:03 +0000 |
| Message-ID | <lu9gu8$847$1@dont-email.me> |
| In reply to | #11926 |
Tim Watts <tw_usenet@dionic.net> wrote: > On 04/09/14 10:02, Aragorn wrote: > > system logs in a binary format, > Do they think they're Microsoft? Even more idiot points for that. The whole of systemd looks like an attempt to bring the microsoft brain damage of a single huge monolith that does everything for everyone all the time to the Linux/Unix world. Apparently Pottering et. al. believe that the braindamaged MS way is "the one true way".
[toc] | [prev] | [next] | [standalone]
| From | Aragorn <thorongil@telenet.be.invalid> |
|---|---|
| Date | 2014-09-04 14:22 +0200 |
| Message-ID | <lu9lha$c81$1@dont-email.me> |
| In reply to | #11930 |
On Thursday 04 September 2014 13:03, Rich conveyed the following to
comp.os.linux.misc...
> Tim Watts <tw_usenet@dionic.net> wrote:
>
>> On 04/09/14 10:02, Aragorn wrote:
>>
>>> system logs in a binary format,
>
>> Do they think they're Microsoft? Even more idiot points for that.
>
> The whole of systemd looks like an attempt to bring the microsoft
> brain damage of a single huge monolith that does everything for
> everyone all the time to the Linux/Unix world.
>
> Apparently Pottering et. al. believe that the braindamaged MS way is
> "the one true way".
I have for a long time already been saying that the latest generation of
GNU/Linux developers are not people who've been involved with UNIX
earlier, but people who come from the Windows world and have only
recently converted to GNU/Linux.
We see this, among other things, in the way that freedesktop.org has
dictated the rules for file managers, for instance. They now refer to
directories as "folders" - in just about every modern-day desktop
environment - and they insist on a volume-oriented approach to storage
via the "Places" menu in the file managers, and which was also already
present in GNOME 2. This volume-oriented approach to storage instead of
the more logical uniform directory tree of UNIX is not exclusive to
Microsoft Windows, of course, because Miguel de Icaza, who designed
GNOME (and who moonlights for Microsoft) is actually a Mac guy.
The late Edsger Dijkstra once said...
"Students who've been exposed to BASIC are brain-damaged beyond
repair."
One could extend this and replace "BASIC" with "Microsoft Windows" in
the above sentence, and it would probably still be correct.
--
= Aragorn =
http://www.linuxcounter.net - registrant #223157
[toc] | [prev] | [next] | [standalone]
| From | Tim Watts <tw_usenet@dionic.net> |
|---|---|
| Date | 2014-09-04 15:06 +0100 |
| Message-ID | <k88mdb-dvd.ln1@squidward.local.dionic.net> |
| In reply to | #11935 |
On 04/09/14 13:22, Aragorn wrote: > I have for a long time already been saying that the latest generation of > GNU/Linux developers are not people who've been involved with UNIX > earlier, but people who come from the Windows world and have only > recently converted to GNU/Linux. At least Imperial College London still teach their CompSci course based primarily on linux PCs in the labs, so there's still hope :) > We see this, among other things, in the way that freedesktop.org has > dictated the rules for file managers, for instance. They now refer to > directories as "folders" - in just about every modern-day desktop > environment - and they insist on a volume-oriented approach to storage > via the "Places" menu in the file managers, and which was also already > present in GNOME 2. This volume-oriented approach to storage instead of > the more logical uniform directory tree of UNIX is not exclusive to > Microsoft Windows, of course, because Miguel de Icaza, who designed > GNOME (and who moonlights for Microsoft) is actually a Mac guy. > I was exposed to volumes on DEC VAX/VMS at around the same time I saw my first unix. Both seemed sensible (notwithstanding MS-DOS's stupid drive letters), but overall I think the unix way is neater.
[toc] | [prev] | [next] | [standalone]
| From | gazelle@shell.xmission.com (Kenny McCormack) |
|---|---|
| Date | 2014-09-04 12:26 +0000 |
| Message-ID | <lu9lqb$kqb$1@news.xmission.com> |
| In reply to | #11930 |
In article <lu9gu8$847$1@dont-email.me>, Rich <rich@example.invalid> wrote: >Tim Watts <tw_usenet@dionic.net> wrote: >> On 04/09/14 10:02, Aragorn wrote: > >> > system logs in a binary format, > >> Do they think they're Microsoft? Even more idiot points for that. > >The whole of systemd looks like an attempt to bring the microsoft brain >damage of a single huge monolith that does everything for everyone all >the time to the Linux/Unix world. > >Apparently Pottering et. al. believe that the braindamaged MS way is >"the one true way". > +1 Clearly, this is what this is all about. The thing that I have been saying for years now, ever since Linux started becoming "user-friendly" and focussing on "big, complex desktops" (read: KDE, GNOME, Unity, etc) is: If I want Windows, I know where to find it. Clearly, "systemd" is all about becoming Windows-like. Red Hat has always been the Microsoft of the Linux world. Note: At the technical level (i.e., not so much at the psychological level), maybe a better analogy is Mac OSX - which is Unix at the core, but with 17 layers of other stuff on top. -- Just for a change of pace, this sig is *not* an obscure reference to comp.lang.c...
[toc] | [prev] | [next] | [standalone]
| From | Tim Watts <tw_usenet@dionic.net> |
|---|---|
| Date | 2014-09-04 15:09 +0100 |
| Message-ID | <9e8mdb-b2e.ln1@squidward.local.dionic.net> |
| In reply to | #11936 |
On 04/09/14 13:26, Kenny McCormack wrote: > Red Hat has always been the Microsoft of the Linux world. I went off RedHat around the time I took over the kerberos side at a uni. Needed to patch the RPMs so I started with their source RPM. Strewth - it had something like 90 patches in it - about half of which were redundant (patch corrects problem that was eventually fixed properly upstream in the current source file). Took me about 3 days to pare it down to something sane. Then there were the lib dependencies...
[toc] | [prev] | [next] | [standalone]
| From | Joe Beanfish <joebeanfish@nospam.duh> |
|---|---|
| Date | 2014-09-05 13:15 +0000 |
| Message-ID | <lucd1p$9sl$2@dont-email.me> |
| In reply to | #11940 |
On Thu, 04 Sep 2014 15:09:45 +0100, Tim Watts wrote: > On 04/09/14 13:26, Kenny McCormack wrote: >> Red Hat has always been the Microsoft of the Linux world. > > I went off RedHat around the time I took over the kerberos side at a > uni. > > Needed to patch the RPMs so I started with their source RPM. > > Strewth - it had something like 90 patches in it - about half of which > were redundant (patch corrects problem that was eventually fixed > properly upstream in the current source file). That's called back-porting. The idea is to fix bugs and holes without changing the feature set and maintain consistent behavior throughout a major rev's updates. > > Took me about 3 days to pare it down to something sane. Then there were > the lib dependencies...
[toc] | [prev] | [next] | [standalone]
| From | Tim Watts <tw_usenet@dionic.net> |
|---|---|
| Date | 2014-09-05 15:16 +0100 |
| Message-ID | <u6todb-gei.ln1@squidward.local.dionic.net> |
| In reply to | #11953 |
On 05/09/14 14:15, Joe Beanfish wrote: > On Thu, 04 Sep 2014 15:09:45 +0100, Tim Watts wrote: > >> On 04/09/14 13:26, Kenny McCormack wrote: >>> Red Hat has always been the Microsoft of the Linux world. >> >> I went off RedHat around the time I took over the kerberos side at a >> uni. >> >> Needed to patch the RPMs so I started with their source RPM. >> >> Strewth - it had something like 90 patches in it - about half of which >> were redundant (patch corrects problem that was eventually fixed >> properly upstream in the current source file). > > That's called back-porting. The idea is to fix bugs and holes without > changing the feature set and maintain consistent behavior throughout > a major rev's updates. You missed the bit about "redundant"... Really - there were things like a 5 line patch that said "fixes <someissue> and then I looked at the source changelog and it said "fixed <someissue>, <somedate after the patch>" Only the upstream hadn't applied the patch - they'd addressed the issue in another part of the code perhaps more correctly - and thus the patch still applied and probably never had any effect after that. I call it sloppy package maintenance.
[toc] | [prev] | [next] | [standalone]
| From | Dan Espen <despen@verizon.net> |
|---|---|
| Date | 2014-09-04 10:10 -0400 |
| Message-ID | <lu9rs8$kn4$1@dont-email.me> |
| In reply to | #11930 |
Rich <rich@example.invalid> writes: > Tim Watts <tw_usenet@dionic.net> wrote: >> On 04/09/14 10:02, Aragorn wrote: > >> > system logs in a binary format, > >> Do they think they're Microsoft? Even more idiot points for that. > > The whole of systemd looks like an attempt to bring the microsoft brain > damage of a single huge monolith that does everything for everyone all > the time to the Linux/Unix world. > > Apparently Pottering et. al. believe that the braindamaged MS way is > "the one true way". I find this unseemly. If you want to snipe at some developer, do it to his face. -- Dan Espen
[toc] | [prev] | [next] | [standalone]
| From | Aragorn <thorongil@telenet.be.invalid> |
|---|---|
| Date | 2014-09-04 14:15 +0200 |
| Message-ID | <lu9l5e$ae5$1@dont-email.me> |
| In reply to | #11926 |
On Thursday 04 September 2014 12:13, Tim Watts conveyed the following to
comp.os.linux.misc...
>> b. systemd has been made into a hard dependency of GNOME/GDM as of
>> GNOME 3.8, which, like systemd, is developed over at RedHat
>> [1]. At the moment and barring more recent information, I
>> believe that this doesn't require systemd to be running as PID
>> 1 yet, but GNOME/GDM does now require systemd to be _installed_
>> either way.
>
> That should have been stopped at birth and is one of the reasons I use
> xfce4 (s/xfce4/your-fav-lightweight-win-manager/)
>
> Both Gnome and KDE have been buggers for innovating in the wrong place
> and putting features in the WM/DE that would have been far more
> sensibly made into a little clean daemon which could interact with any
> desktop (or none at all for terminal users) via a common messaging
> interface (which is dbus I guess).
Well, I personally like KDE - I've been using it since version 1.1.1, in
my first ever GNU/Linux distribution, Mandrake 6.0 PowerPack - but in
all fairness, KDE is trying to comply with the rules set forth by
freedesktop.org. And that's RedHat. Who develop GNOME.
The Nepomuk in KDE was something proposed at the level of the EU, and
KDE have adopted that, but they are moving away from it again at the
moment, as I understand it.
As for a messaging bus, KDE had dcop, but considering that this was a
KDE innovation - read: not RedHat-originated [*] - RedHat created dbus
as an alternative.
[*] It is no secret within the community that RedHat hates KDE. Even
long after KDE had already started using the freely licensed Qt2
widgets - the original KDE 1 used non-free Qt1 libraries - RedHat
kept on carrying on with a hate campaign against KDE and its
developers, leading RedHat to be named HatRed by some.
> Taken to extremes, this approach would have had us only able to run
> KDE apps on a KDE DE and Gnome apps on Gnome. Silly example but it
> does follow as a logical extreme.
Well, KDE is definitely not shy of interoperability and the KDE
developers have tried to play ball with everyone, but the truth of the
matter is that the GNOME developers looked at KDE and then tried
reinventing every wheel for their own purposes. It is in fact GNOME and
the GTK developers who are very strongly anti-interoperable and who
insist on doing things their way, and their way only.
> Fortunately, the greater good prevailed there.
>
>> systemd has brought a number of innovations and good ideas to
>> GNU/Linux, but the whole concept has been poorly implemented from the
>> start, and along with those good ideas and innovations also came a
>> great number of bad ideas, such as
>
>> "the /usr move" [2],
>
> Oh god - I don't even know what that is...
It is comprised of moving all runtime binaries off of the root
filesystem and into /usr, with /bin, /sbin and /lib{,64} only being
symbolic links to their respective equivalents under /usr.
This means that when /usr is on a separate filesystem, then one can no
longer boot without an initramfs - typically created using dracut -
which mounts /usr before systemd/udev is started. (udev is now
integrated with systemd, which is why some of the Gentoo developers
created eudev as a fork from an earlier udev version.)
>> a 1 MiB monolith as PID 1,
>
> That's idiotic and kills it for embedded systems.
Exactly.
>> system logs in a binary format,
>
> Do they think they're Microsoft? Even more idiot points for that.
>
>> and very sloppy coding, to say the very least.
>
> And a poor attitude to bug reports as per my last post.
Yes, I had not seen those particular exchanges yet myself, but I was
already aware of Sievers' and Poettering's arrogance, and I
wholeheartedly agree.
>> systemd violates the UNIX philosophy of having small, distinct and
>> interoperative tools that do one thing but do it excellently.
>> There's a good reason as to why UNIX has survived for over 40 years:
>> it works and it's easily maintainable. The arrogance displayed by
>> Sievers and Poettering [3] is a different matter altogether, but
>> nevertheless doesn't help in remedying the situation.
>
> For me, I will be putting it in the same can as Wayland - a cute idea
> at its core, but so far off the mark in practice I am not touching it
> with a bargepole.
I share your sentiments.
>> In addition to all of this, the traditional System V init was never
>> broken. Embedded devices or things that need to be booted up
>> multiple times a day can benefit from the marginal increase in boot
>> time offered by systemd, but systems that remain up 24/7 don't care
>> about boot times, and especially not if those boot times are only
>> marginally faster with systemd than they are with the traditional
>> System V init.
>
> Indeed. Full agreement.
>
>>
>> [1] It is obviously a strategic move by RedHat, who are trying to
>> usurp the market of enterprise-oriented GNU/Linux distributors,
>> at the cost of flexibility and choice towards the users and
>> staying with the proven UNIX tradition. The same attitude can be
>> seen among the GNOME and even the GTK developers.
>>
>> [2] While the current argument for "the /usr move" - i.e. that with
>> all binaries under /usr, it would be easier to have /usr mounted
>> read- only - makes sense, that was not the reason behind it in
>> the first place. The reason was that Poettering and Sievers had
>> screwed up and put boot-essential stuff under /usr - where it
>> doesn't belong - so as to placate a few corner cases.
>
> Foolish... There have been occasions when I have had to rescue
> something with a broken /usr/ (disk problem).
The current philosophy of the systemd developers is that the initramfs
can provide for a rescue shell for those particular instances.
Distributions typically set up dracut to provide for that via busybox.
>> [3] "We are RedHat. Resistance is futile. You will be assimilated."
>> Lennart Poettering has more than once stated on his blog that he
>> joyfully looks forward to the day where no other init systems
>> than systemd will exist anymore, and has also equally arrogantly
>> mocked other distributions - among whom Gentoo - on his blog.
>
> Redhat might still have a certain following[1], but in my world of
> linux in the UK I see a *lot* of debian and ubuntu deployment - both
> in academia and in startup tech companies in London.
>
> [1] Notably in banks and CERN/High energy physics as Scientific Linux
> is CERN's supported flavour and that's based off RHEL/Centos and
> originates from FermiLab
Yes, I'm aware of Scientific Linux, and it's a long term support
distribution, so it's still safe for a while, but eventually it too will
be yielding to the transition to systemd. It has no choice, being based
upon RedHat.
> The only thing Centos/RHEL is good for is its hugely long support
> lifetimes of certain releases. But I persoanally find it very
> frustrating to work with compared to Debian Stable. In fact I've even
> gone to Debian on my desktop (but Mint on my laptop). It takes a
> little more effort to polish it up but not much...
My former colleagues and I used to use CentOS on our servers back when
we were still running an IRC network, because the hosting software we
were using only supported CentOS or RedHat. However, I agree; I too
would rather use Debian Stable.
[toc] | [prev] | [next] | [standalone]
| From | Tim Watts <tw_usenet@dionic.net> |
|---|---|
| Date | 2014-09-04 15:03 +0100 |
| Message-ID | <g28mdb-qsd.ln1@squidward.local.dionic.net> |
| In reply to | #11934 |
On 04/09/14 13:15, Aragorn wrote:
> As for a messaging bus, KDE had dcop, but considering that this was a
> KDE innovation - read: not RedHat-originated [*] - RedHat created dbus
> as an alternative.
>
> [*] It is no secret within the community that RedHat hates KDE. Even
> long after KDE had already started using the freely licensed Qt2
> widgets - the original KDE 1 used non-free Qt1 libraries - RedHat
> kept on carrying on with a hate campaign against KDE and its
> developers, leading RedHat to be named HatRed by some.
Or DeadRat by me.
>> Taken to extremes, this approach would have had us only able to run
>> KDE apps on a KDE DE and Gnome apps on Gnome. Silly example but it
>> does follow as a logical extreme.
>
> Well, KDE is definitely not shy of interoperability and the KDE
> developers have tried to play ball with everyone, but the truth of the
> matter is that the GNOME developers looked at KDE and then tried
> reinventing every wheel for their own purposes. It is in fact GNOME and
> the GTK developers who are very strongly anti-interoperable and who
> insist on doing things their way, and their way only.
Perhaps I unfairly lumped KDE in there - I often preferred the KDE apps,
but not the desktop as it seemed very "big" and prone to going doo-lally
if one of the sub components like DCOP failed. DCOP seemed to have
issues with multiple concurrent logins on different PCs sharing an NFS
$HOME - but this was 2005 - maybe that's better now?)
With XFCE4 of course (and many other smaller WMs) you can restart
components without blowing all your apps away and the WM/DE components
don't depend on much.
But I did like the KDE "clean presentation" - always seemed very slick
and artistically designed. If there was a "KDE-lite" WM/DE I would
probably use it - if that even makes sense!
>>> "the /usr move" [2],
>>
>> Oh god - I don't even know what that is...
>
> It is comprised of moving all runtime binaries off of the root
> filesystem and into /usr, with /bin, /sbin and /lib{,64} only being
> symbolic links to their respective equivalents under /usr.
And since when did the init process dictate FS naming policy?
I'm not even sure why they would want to do that or why anyone would
think it a good idea.
> This means that when /usr is on a separate filesystem, then one can no
> longer boot without an initramfs - typically created using dracut -
> which mounts /usr before systemd/udev is started. (udev is now
> integrated with systemd, which is why some of the Gentoo developers
> created eudev as a fork from an earlier udev version.)
>
> Yes, I had not seen those particular exchanges yet myself, but I was
> already aware of Sievers' and Poettering's arrogance, and I
> wholeheartedly agree.
The odd arrogant outburst is tolerable if you are mostly usually right
plus actually being very very smart (eg Linus, djb). But in this case, hmmm.
>
> The current philosophy of the systemd developers is that the initramfs
> can provide for a rescue shell for those particular instances.
> Distributions typically set up dracut to provide for that via busybox.
I see where they are coming from, but I personally prefer my rescue
shell to be "init=/bin/bash" and a working / where possible (sans /usr
if needs be).
> Yes, I'm aware of Scientific Linux, and it's a long term support
> distribution, so it's still safe for a while, but eventually it too will
> be yielding to the transition to systemd. It has no choice, being based
> upon RedHat.
Unless someone forks RHEL :)
>> The only thing Centos/RHEL is good for is its hugely long support
>> lifetimes of certain releases. But I persoanally find it very
>> frustrating to work with compared to Debian Stable. In fact I've even
>> gone to Debian on my desktop (but Mint on my laptop). It takes a
>> little more effort to polish it up but not much...
>
> My former colleagues and I used to use CentOS on our servers back when
> we were still running an IRC network, because the hosting software we
> were using only supported CentOS or RedHat. However, I agree; I too
> would rather use Debian Stable.
>
[toc] | [prev] | [next] | [standalone]
| From | Benjamin Rochefort <ben@benwen.info> |
|---|---|
| Date | 2014-09-04 20:23 -0400 |
| Message-ID | <luavqh$fnt$1@speranza.aioe.org> |
| In reply to | #11938 |
Le 2014-09-04 10:03, Tim Watts a écrit : > But I did like the KDE "clean presentation" - always seemed very slick > and artistically designed. If there was a "KDE-lite" WM/DE I would > probably use it - if that even makes sense! LXDE and Razor-qt merged recently and the resulting desktop environment may be the 'KDE-lite' DE you're looking for!
[toc] | [prev] | [next] | [standalone]
| From | Aragorn <thorongil@telenet.be.invalid> |
|---|---|
| Date | 2014-09-05 06:41 +0200 |
| Message-ID | <lubess$ja1$1@dont-email.me> |
| In reply to | #11948 |
On Friday 05 September 2014 02:23, Benjamin Rochefort conveyed the
following to comp.os.linux.misc...
> Le 2014-09-04 10:03, Tim Watts a écrit :
>
>> But I did like the KDE "clean presentation" - always seemed very
>> slick and artistically designed. If there was a "KDE-lite" WM/DE I
>> would probably use it - if that even makes sense!
>
> LXDE and Razor-qt merged recently and the resulting desktop
> environment may be the 'KDE-lite' DE you're looking for!
And then there's also still Trinity Desktop Environment, which is a fork
of KDE 3.5. Many distributions have it in their repos, and for those
that don't, the TDE website itself may offer compatible packages. ;-)
http://www.trinitydesktop.org
--
= Aragorn =
http://www.linuxcounter.net - registrant #223157
[toc] | [prev] | [next] | [standalone]
| From | Joe Beanfish <joebeanfish@nospam.duh> |
|---|---|
| Date | 2014-09-05 13:23 +0000 |
| Message-ID | <lucdfn$9sl$3@dont-email.me> |
| In reply to | #11938 |
On Thu, 04 Sep 2014 15:03:28 +0100, Tim Watts wrote:
> On 04/09/14 13:15, Aragorn wrote:
>>>> "the /usr move" [2],
>>>
>>> Oh god - I don't even know what that is...
>>
>> It is comprised of moving all runtime binaries off of the root
>> filesystem and into /usr, with /bin, /sbin and /lib{,64} only being
>> symbolic links to their respective equivalents under /usr.
>
> And since when did the init process dictate FS naming policy?
>
> I'm not even sure why they would want to do that or why anyone would
> think it a good idea.
I'm only guessing at their rationale here but...
It's common these days for /usr to be the same FS as / because of large
disks and no longer needing a tiny / for bootstrap. In that case it's
a bit pointless to have redundant copies of stuff in /bin and /usr/bin
etc..
[toc] | [prev] | [next] | [standalone]
| From | William Unruh <unruh@invalid.ca> |
|---|---|
| Date | 2014-09-05 13:47 +0000 |
| Message-ID | <lucess$p00$1@dont-email.me> |
| In reply to | #11954 |
On 2014-09-05, Joe Beanfish <joebeanfish@nospam.duh> wrote:
> On Thu, 04 Sep 2014 15:03:28 +0100, Tim Watts wrote:
>> On 04/09/14 13:15, Aragorn wrote:
>>>>> "the /usr move" [2],
>>>>
>>>> Oh god - I don't even know what that is...
>>>
>>> It is comprised of moving all runtime binaries off of the root
>>> filesystem and into /usr, with /bin, /sbin and /lib{,64} only being
>>> symbolic links to their respective equivalents under /usr.
>>
>> And since when did the init process dictate FS naming policy?
>>
>> I'm not even sure why they would want to do that or why anyone would
>> think it a good idea.
>
> I'm only guessing at their rationale here but...
They made more and more stuff in the boot process dependent on libraries
and programs in /usr/{bin,lib} rather than just in /lib. So one either
moved all of those things into /{bin,lib} or left them where they were
and used /usr/{bin,lib} The problem is that lots of people have /usr on
a different partition from /. So you must eitehr demand that /usr be
mounted in initrd or that it be onthe same partition. Once you had done
that then maintianing separate /{bin,lib} from /usr/{bin,lib} became
pointless. The post facto rational was what you list below, but that was
certainly not the reason. It was that the whole push to udev/systemd had
become a {fat pig,feature laden} which depended on far to much userland stuff.
>
> It's common these days for /usr to be the same FS as / because of large
> disks and no longer needing a tiny / for bootstrap. In that case it's
> a bit pointless to have redundant copies of stuff in /bin and /usr/bin
> etc..
[toc] | [prev] | [next] | [standalone]
| From | Tim Watts <tw_usenet@dionic.net> |
|---|---|
| Date | 2014-09-05 15:26 +0100 |
| Message-ID | <0qtodb-lti.ln1@squidward.local.dionic.net> |
| In reply to | #11955 |
On 05/09/14 14:47, William Unruh wrote:
> They made more and more stuff in the boot process dependent on libraries
> and programs in /usr/{bin,lib} rather than just in /lib. So one either
> moved all of those things into /{bin,lib} or left them where they were
> and used /usr/{bin,lib} The problem is that lots of people have /usr on
> a different partition from /. So you must eitehr demand that /usr be
> mounted in initrd or that it be onthe same partition. Once you had done
> that then maintianing separate /{bin,lib} from /usr/{bin,lib} became
> pointless. The post facto rational was what you list below, but that was
> certainly not the reason. It was that the whole push to udev/systemd had
> become a {fat pig,feature laden} which depended on far to much userland stuff.
That does seem to explain it very well.
Though I cannot see exactly where the deps are.
[toc] | [prev] | [next] | [standalone]
| From | Tim Watts <tw_usenet@dionic.net> |
|---|---|
| Date | 2014-09-05 15:17 +0100 |
| Message-ID | <m9todb-gei.ln1@squidward.local.dionic.net> |
| In reply to | #11954 |
On 05/09/14 14:23, Joe Beanfish wrote:
> On Thu, 04 Sep 2014 15:03:28 +0100, Tim Watts wrote:
>> On 04/09/14 13:15, Aragorn wrote:
>>>>> "the /usr move" [2],
>>>>
>>>> Oh god - I don't even know what that is...
>>>
>>> It is comprised of moving all runtime binaries off of the root
>>> filesystem and into /usr, with /bin, /sbin and /lib{,64} only being
>>> symbolic links to their respective equivalents under /usr.
>>
>> And since when did the init process dictate FS naming policy?
>>
>> I'm not even sure why they would want to do that or why anyone would
>> think it a good idea.
>
> I'm only guessing at their rationale here but...
>
> It's common these days for /usr to be the same FS as / because of large
> disks and no longer needing a tiny / for bootstrap. In that case it's
> a bit pointless to have redundant copies of stuff in /bin and /usr/bin
> etc..
>
There wouldn't be redundant copies though - stuff that's in /bin stays
there and isn't replicated to /usr/bin as both are in $PATH for any
given process.
I really don't get it ---
[toc] | [prev] | [next] | [standalone]
| From | William Unruh <unruh@invalid.ca> |
|---|---|
| Date | 2014-09-05 20:51 +0000 |
| Message-ID | <lud7nk$67o$2@dont-email.me> |
| In reply to | #11958 |
On 2014-09-05, Tim Watts <tw_usenet@dionic.net> wrote:
> On 05/09/14 14:23, Joe Beanfish wrote:
>> On Thu, 04 Sep 2014 15:03:28 +0100, Tim Watts wrote:
>>> On 04/09/14 13:15, Aragorn wrote:
>>>>>> "the /usr move" [2],
>>>>>
>>>>> Oh god - I don't even know what that is...
>>>>
>>>> It is comprised of moving all runtime binaries off of the root
>>>> filesystem and into /usr, with /bin, /sbin and /lib{,64} only being
>>>> symbolic links to their respective equivalents under /usr.
>>>
>>> And since when did the init process dictate FS naming policy?
>>>
>>> I'm not even sure why they would want to do that or why anyone would
>>> think it a good idea.
>>
>> I'm only guessing at their rationale here but...
>>
>> It's common these days for /usr to be the same FS as / because of large
>> disks and no longer needing a tiny / for bootstrap. In that case it's
>> a bit pointless to have redundant copies of stuff in /bin and /usr/bin
>> etc..
>>
>
> There wouldn't be redundant copies though - stuff that's in /bin stays
> there and isn't replicated to /usr/bin as both are in $PATH for any
> given process.
Nope on the more recent systems, everything in /bin has been moved to
/usr/bin, and /bin made a link to /usr/bin.
>
> I really don't get it ---
[toc] | [prev] | [next] | [standalone]
Page 7 of 10 — ← Prev page 1 … 5 6 [7] 8 9 10 Next page →
Back to top | Article view | comp.os.linux.misc
csiph-web