On Sun, Dec 1, 2013 at 12:33 PM, Kay Sievers <[email protected]> wrote: > On Sun, Dec 1, 2013 at 8:25 PM, Hristo Venev <[email protected]> wrote: >> I've implemented session mode for systemd. I am currently using it and it >> works >> pretty okay. I just had to generate D-Bus user services for D-Bus activation >> to >> work. >> >> A new variable, XDG_SESSION_DIR, has been added. It defaults to >> /run/session/$XDG_SESSION_ID and is to be used for session-specific files as >> it >> is `rm -rf`-ed on logout. There, in ./systemd/private, the systemd socket is >> stored. Also, my implementation of session units stores the D-Bus socket in >> ./dbus/session_bus_socket. > > There is intentionally only support for "systemd --user", no plan to > support any sort of systemd session instance. The one instance for the > user will be shared by all sessions of the same user.
right, my personal thoughts on this were to have session@<seat>.service units that take care of spawing a graphical UX to the user on that seat. I see little reason to create yet another instance of systemd --user as the one existing process should be able to handle everything for a single user already. Not sure what to do with this patch series, I fear it will risk running into problems when you have multiple seats and other things like that, but I haven't looked into details of the patches too much... Auke _______________________________________________ systemd-devel mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/systemd-devel
