[systemd-devel] user units and system units behavior

[Simon McVittie]

On 14/02/15 18:26, Ivan Shapovalov wrote:
> Yes, the per-session bus is there, but it is not used at all for
> communication with per-user systemd instance.

I do want this to work, and I'm working on making it happen. It works on 
my Debian system, with the patched dbus that I recently uploaded to 
experimental.

When my patches on https://bugs.freedesktop.org/show_bug.cgi?id=61301 
have been merged, if dbus is compiled with --enable-user-session and you 
are running systemd, the per-login-session "dbus-daemon --session" will 
be replaced by a per-user-session "dbus-daemon --session" (see earlier 
thread for explanation of login session vs. user session). At that 
point, the "dbus-daemon --session" can be suitable for communicating 
with `systemd --user`.

(I believe the plan is (still) that kdbus systems will always have an 
equivalent of this new per-user-session bus, and never a 
per-login-session bus.)

> No, mine /etc/X11/xinitrc.d is Simon's /etc/X11/Xsession.d and "similar
> setups". It's apparently a distro-specific path.

Yes. I think /etc/X11/xinitrc.d is what Red Hat and its derivatives use. 
Xsession.d is used instead in Debian and its derivatives, including 
Ubuntu. The differences are because, historically, this sort of plumbing 
was something that every distribution had to invent for itself.

     S

-- 
Simon McVittie
Collabora Ltd. <http://www.collabora.com/>