[Pkg-utopia-maintainers] Bug#520720: Bug#520720: gnome-power-manager: Stopped working after consolekit upgrade)

[Sam Morris]

On Tue, 2009-03-24 at 19:19 +0100, Michael Biebl wrote:
> Sam Morris wrote:
> >> what's happening (afaics) is this.
> >>
> >> 1.) you have libpam-ck-connector installed, which with version 0.3.0-1 is now
> >> automatically enabled (using pam-auth-update)
> >> 2.) When you login on the console, XDG_SESSION_COOKIE will be set by the pam module
> >> 3.) you then start your X session via startx, the Xsession script will not run
> >> though, as XDG_SESSION_COOKIE is already set.
> >> 4.) Your session on the console (tty1) will then correctly be marked as inactive
> >> (as your active session will now be at tty7)
> > 
> > Ah,  I see, that all makes sense now!
> > 
> >> For the time being, you can either uninstall libpam-ck-connector / disable it
> >> (run dpkg-reconfigure libpam-ck-connector) or remove the XDG_SESSION_COOKIE
> >> check in /etc/X11/Xsession.d/90consolekit (which shouldn't do any harm for you,
> >> as you don't use gdm/kdm)
> >>
> >> I'm not quite sure yet, how to address this issue.
> > 
> > Do you know what the harmful side-effects of removing the
> > XDG_SESSION_COOKIE check in 90consolekit are? I don't really see what
> > the point of the check is, and it breaks startx & friends.
> 
> gdm/kdm talk to consolekit directly.
> So if you login via a login manager (which also launches the xsession scripts),
> you would get two consolekit session cookies.

Ugh, I see. :(

Is there any advantage in having display managers talk to consolekit
directly? It seems like duplication of effort given that
libpam-ck-connector exists.

Alternatively, what do you think about having 90consolekit test for
well-known environment variables rather than XDG_SESSION_COOKIE? I see
that GDM sets GDMSESSION... if the other consolekit-using display
managers set other variables, the script could test for the existence of
any of them.

I guess it would be better for all the ConsoleKit-using display managers
to, if they insist on not relying on libpam-ck-connector, to set a
common environment variable that the script can test for.

> Cheers,
> Michael

-- 
Sam Morris
https://robots.org.uk/

PGP key id 1024D/5EA01078
3412 EA18 1277 354B 991B  C869 B219 7FDB 5EA0 1078
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part
Url : http://lists.alioth.debian.org/pipermail/pkg-utopia-maintainers/attachments/20090324/46cf09c3/attachment-0001.pgp