From: Koen Kooi <koen@dominion.thruhere.net>
To: openembedded-devel@lists.openembedded.org
Subject: Re: [meta-oe][RFC 00/27] systemd / initmanager rework
Date: Wed, 08 Feb 2012 14:05:28 +0100 [thread overview]
Message-ID: <jgtruq$6gb$1@dough.gmane.org> (raw)
In-Reply-To: <20120208124311.GB3647@jama.jama.net>
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Op 08-02-12 13:43, Martin Jansa schreef:
> On Wed, Feb 08, 2012 at 10:28:12AM -0200, Otavio Salvador wrote:
>> On Wed, Feb 8, 2012 at 10:14, Koen Kooi <koen@dominion.thruhere.net>
>> wrote:
>>
>>> Images with sysvinit will be working the same as before, just with
>>> libsystemd-logind.so.X being installed. Images with systemd get
>>> working session tracking.
>>>
>>
>> AFAIK polkit uses consolekit or systemd. In case you build it with
>> systemd you break consolekit support.
>>
>> However a stronger point IMO is that some changes, for one or another
>> init system, might involve specific compilation options and thus making
>> their runtime incompatible as it's the case of polkit. Usually people
>> will want one or another init system and I see no gain in have both
>> available.
>>
>> Forcing user to use a bbappend to have one working well is the wrong
>> way.
>
> I think that only usable alternative _if it works_ is to build polkit
> linked always to libsystemd-logind.so.X and force users of sysvinit to
> use small part of systemd (libsystemd-logind.so.X packaged in
> systemd-logind).
>
> Then we could keep add --enable-systemd to polkit etc., but if this
> breaks stuff which worked with consolekit, then distro policy would be
> better.
This is not really directed at Martin, but at the wider audience:
For the past 6 months I have been trying to make consolekit work:
http://cgit.openembedded.org/openembedded-core/commit/meta/recipes-support/consolekit?id=eb828308b8e721e4ad7edbd15f47ecad22ee81cf
http://cgit.openembedded.org/openembedded-core/commit/meta/recipes-support/consolekit?id=5608a748af2c754f60137ab7c3010ccce6bf9e40
And right now it still isn't working properly. Look real close at the first
commit; oe-core developers made a serious effort to make it
non-functional[1], only linking needed to succeeded. So please stop saying
CK will break, it already is broken.
Now back to the original discussion: support setting multiple init managers
in DISTRO_FEATURES and keep the final choice an IMAGE_FEATURE. The end
result: the integrator keeps the same freedoms and when having only a single
init manager in DISTRO_FEATURES things will slightly improve.
regards,
Koen
[1] To be fair, that was not their goal, but it was the end result. No
malice there, I verified :)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (Darwin)
Comment: GPGTools - http://gpgtools.org
iEYEARECAAYFAk8ycxMACgkQMkyGM64RGpF/MwCfdLEnjLDnZDtN5VhPt0mDZRzi
LbcAn2aYuUZXcPkMRY2qfMYK2dgmSTC5
=sX7J
-----END PGP SIGNATURE-----
next prev parent reply other threads:[~2012-02-08 13:13 UTC|newest]
Thread overview: 66+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-02-07 15:12 [meta-oe][RFC 00/27] systemd / initmanager rework Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 01/27] systemd.bbclass: rework Andreas Müller
2012-02-09 8:14 ` Andreas Müller
2012-02-09 11:03 ` Otavio Salvador
2012-02-07 15:12 ` [meta-oe][RFC 02/27] initmanager.bbclass: inital add Andreas Müller
2012-02-07 15:39 ` Paul Eggleton
2012-02-07 15:57 ` Otavio Salvador
2012-02-07 22:53 ` Khem Raj
2012-02-07 15:12 ` [meta-oe][RFC 03/27] image.bbclass: inherit initmanager.bbclass and set default value for IMAGE_INIT_MANAGER Andreas Müller
2012-02-07 15:36 ` Paul Eggleton
2012-02-07 15:59 ` Otavio Salvador
2012-02-07 15:12 ` [meta-oe][RFC 04/27] elsa: remove unneeded systemd code Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 05/27] connman: " Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 06/27] openssh: " Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 07/27] gateone: " Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 08/27] cronie: " Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 09/27] xserver-nodm-init: " Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 10/27] busybox: " Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 11/27] atftp: " Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 12/27] gpsd: " Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 13/27] gdm: inherit systemd / " Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 14/27] dropbear: " Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 15/27] cloud9: pack systemd support in ${PN}-systemd / " Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 16/27] slim: inherit systemd " Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 17/27] udisks: conditional DEPENDS " Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 18/27] rsyslog: " Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 19/27] polkit: conditional DEPEND systemd Andreas Müller
2012-02-07 15:35 ` Martin Jansa
2012-02-07 15:40 ` Paul Eggleton
2012-02-07 15:12 ` [meta-oe][RFC 20/27] syslog-ng: conditional --enable-systemd --with-systemdsystemunitdir / remove unneeded systemd code Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 21/27] lighthttpd: inherit systemd / added systemd native support " Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 22/27] cherokee: " Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 23/27] networkmanager: inherit systemd / added NetworkManager-wait-online.service " Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 24/27] ntp: add native systemd support for ntpdate " Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 25/27] dhcp: add native support for dhcp-relay " Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 26/27] task-basic: replace systemd specific packets by common ones / conditionally RDEPENDS avahi(-systemd) systemd-compat-units Andreas Müller
2012-02-07 15:12 ` [meta-oe][RFC 27/27] task-x11: replace systemd specific packets by common ones Andreas Müller
2012-02-07 15:30 ` Martin Jansa
2012-02-07 15:52 ` [meta-oe][RFC 00/27] systemd / initmanager rework Koen Kooi
2012-02-07 16:24 ` Andreas Müller
2012-02-07 17:14 ` Koen Kooi
2012-02-07 17:58 ` Andreas Müller
2012-02-07 15:55 ` Otavio Salvador
2012-02-07 16:02 ` Koen Kooi
2012-02-07 16:18 ` Otavio Salvador
2012-02-08 0:39 ` Joshua Lock
2012-02-08 8:18 ` Koen Kooi
2012-02-08 11:00 ` Otavio Salvador
2012-02-08 11:07 ` Koen Kooi
2012-02-08 11:12 ` Otavio Salvador
2012-02-08 12:14 ` Koen Kooi
2012-02-08 12:18 ` Paul Eggleton
2012-02-08 12:21 ` Koen Kooi
2012-02-08 12:49 ` Koen Kooi
2012-02-08 12:28 ` Otavio Salvador
2012-02-08 12:43 ` Martin Jansa
2012-02-08 13:05 ` Koen Kooi [this message]
2012-02-08 12:51 ` Andreas Müller
2012-02-08 12:54 ` Koen Kooi
2012-02-08 12:56 ` Martin Jansa
2012-02-08 13:19 ` Khem Raj
2012-02-08 15:35 ` Otavio Salvador
2012-02-08 11:31 ` Andreas Müller
2012-02-08 10:57 ` Otavio Salvador
2012-02-07 16:35 ` Andreas Müller
2012-02-07 16:39 ` Otavio Salvador
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='jgtruq$6gb$1@dough.gmane.org' \
--to=koen@dominion.thruhere.net \
--cc=openembedded-devel@lists.openembedded.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox